Year book - Carnegie Institution of Washington

Carnegie
Institution

OF WASHINGTON

.- " v V' i.^!>-''::m:\:Hm;^i^^^

Carnegie
Institution

OF WASHINGTON

Year Book 76

1976-1977

Library of Congress Catalog Card Numh)er 3-16716

Port City Press, Inc., Baltimore, Maryland

Issued December 1977

Contents

Officers and Staff v

Report of the President 1

Reports of Departments and Special Studies 1

Department of Embryology 3

Hale Observatories 115

Department of Plant Biology 199

Geophysical Laboratory 377

Developmental Biology Research Group 693

Department of Terrestrial Magnetism 713

Bibliography 903

Administrative Reports 905

Report of the Executive Committee 907

Abstract of Minutes of the Eightieth Meeting of the

Board of Trustees 909

Financial Statement 911

Articles of Incorporation 929

By-Laws of the Institution 933

Index of Names . 939

111

President and Trustees

PEESIDENT
Philip H. Abelson

BOAPvD OF TRUSTEES

Frank Stanton
Chaii-man

William C. Greenough
Vice-Chaiwian

William T. Golden
Secretary

Kobert 0. Anderson

J. Paul Austin

Lewis M. Branscomb

John T. Connor

John Diebold

Michael Ference, Jr.

Carl J. Gilbert

Hanna H. Gray

Crawford H. Greenewalt

Caryl P. Haskins

William R. Hewlett

William MrChesney Martin, Jr.

Henry S. Morgan

Waher H. Page

Robert M. Pennoyer

Richard S. Perkins

William M. Roth

Robert C. Seamans, Jr.

Charles H. TowTies

Juan T. Trippe

James N. White

Garrison Norton
Charles P. Taft

Trustees Emeriti

Former Presidents and Trustees

PRESIDENTS

Daniel Coit Oilman 1902-1904 John Campbell Merriam

Robert Simpson Woodward 1904-1920 Vannevar Bush

Canl P. Haskins 195&-1971

TRUSTEES

Alexander Agassiz
Lord Ashby of Brandon
George J. Baldwin
Thomas Barbour
James F. Bell
John S. Billings
Robert Woods Bliss
Amory H. Bradford
Lindsay Bradford
Omar N. Bradley
Robert S. Brookings
"N'aimevar Bush
John L. Cadwalader
William W. Campbell
John J. Carty
Whitefoord R. Cole
Frederic A. Delano
Cleveland H. Dodge
William E. Dodge
Charles P. Fenner
Homer L. Ferguson
Simon Flexner
W. Cameron Forbes
James Forrestal
William X. Frew
Lvman J. Gage
Walter S. Gifford
Cass Gilbert
Frederick H. Gillett
Daniel C. Oilman
Patrick E. Haggerty
John Hay

Barklie McKee Henry
Myron T. Herrick
Abram S. Hewitt
Henry L. Higginson
Ethan A. Hitchcock
Henry Hitchcock
Herbert Hoover
William Wirt Howe
Charles L. Hutchinson
Walter A. Jessup
Frank B. Jewett
Samuel P. Langley
Ernest 0. Lawrence
Charles A. Lindbergh
William Lindsay
Henry Cabot Lodge
Alfred L. Loomis

1904-05

1967-74

1925-27

1934-46

1935-61

1902-13

193G-62

1959-72

1940-58

194g-69

1910-29

1958-71

1903-14

1929-38

1916-32

1925-34

1927^9

1903-23

1902-03

1914-24

1927-52

1910-14

1920-55

1948-49

1902-15

1902-12

1931-66

1924-34

1924-35

1902-08

1974-75

190^05

1949-66

1915-29

1902-03

1902-19

1902-09

1902

1920-49

1903-09

1902-04

1938-44

1933-49

1904-06

1944-58

1934-39

1902-09

1914-24

1934-73

Robert A. Lovett
Seth Low
Wavne MacVeagh
Keith S. McHugh
Andrew W. Mellon
John Campbell Merriam
Margaret Carnegie Miller
Roswell Miller
Darius 0. Mills
S. Weir Mitchell
Andrew J. Montague
William W. Morrow
Seeley G. Mudd
William L Myers
William Church Osborn
James Parmelee
Wm. Barclay Parsons
Stewart Paton
George W. Pepper
John J. Pershing
Henning W. Prentis, Jr.
Henry S. Pritchett
Gordon S. Rentschler
David Rockefeller
Elihu Root
Elihu Root, Jr.
Julius Rosenwald
William W. Rubey
Martin A. Ryerson
Henry R. Shepley
Theobald Smith
John C. Spooner
William Benson Storey
Richard P. Strong
William H. Taft
William S. Thayer
James W. Wadsworth
Charles D. Walcott
Frederic C. Walcott
Henry P. Walcott
Lewis H. Weed
William H. Welch
Andrew D. White
Edward D. White
Henry White
George W. Wickersham
Robert E. Wilson
Robert S. Woodward
Carroll D. Wright

1921-1938
1939-1955

1948-71
1902-16
1902-07
1950-74
1924-37
1921-38
1955-67
1933-55
1902-09
1902-14
1907-35
1902-29
1940-68
1948-76
1927-34
1917-31
1907-32
1916-42
1914-19
1930-43
1942-59
1906-36
1946-48
1952-56
1902-37
1937-67
1929-31
1962-74
1908-28
1937-62
1914-34
1902-07
1924-39
1934-48
1906-15
1929-32
1932-52
1902-27
1931-48
1910-24
1935-52
1906-34
1902-16
1902-03
1913-27
1909-36
1953-64
1905-24
1902-08

Under the original charter, from the date of organization until April 28, 1904, the following
were ex officio members of the Board of Trustees: the President of the United States, the Presi-
dent of the Senate, the Si:)eaker of the House of Rei)resentatives, the Secretary of the
Smithsonian Institution, and the President of the National Academy of Sciences.

VI

OFFICE OF ADMINISTRATION

1530 P Street, N.W., Washington, D.C. 20005

Philip H. Abelson President

James W. Boise Bursar; Secretary -Treasurer, Retirement Trust;
Executive Secretary to the Finance Committee

Marjorie H. Walburn Assistant to the President

Sheila A. McGough Publications Officer; Editor

Kenneth R. Henard Assistant Bursar; Assistant Treasurer,

Retirement Trust

Joseph M. S. Haraburda Assistant to the Bursar

Richard E. Hewitt Administrative Officer for Services

Marshall Hornblower Counsel

STAFF MEMBERS IN SPECIAL SUBJECT AREAS

Elhs T. Bolton
Roj^ J. Britten
Tatiana Proskouriakoff^

DISTINGUISHED SERVICE MEMBER IN SPECIAL SUBJECT AREA

Barbara McClintock

RESEARCH ASSOCIATE AT LARGE

Harry E. D. Pollock

1 Retired June 30, 1977.

Vll

Carnegie Institution of Washington adheres in all phases of its operations,
including employment and educational programs, to a policy barring discrimina-
tion on the basis of race, religion, color, national or ethnic origin, sex, or physical
handicap. In its educational programs it admits qualified students as fellows
without regard to race, religion, color, national or ethnic origin, sex, or physical
handicap to all the rights, privileges, programs, and activities generally accorded
or made available to fellows at the Institution. It does not discriminate on the
basis of race, religion, color, national or ethnic origin, sex, or physical handicap
in administration of its educational policies, admissions policies, fellowship
programs, and other Institution-administered programs.

Report OF
THE President

INTRODUCTION

AT THE CARNEGIE INSTITUTION it is ouF custom to engage in frequent
self -appraisal. Every year the professional staff and fellows prepare a
comprehensive report presenting the results of their research and defending
its significance. Over the years a rich record of changing aspirations and
accomplishments has been compiled in these Annual Reports.

Another tradition of the Institution is that from time to time its presi-
dents make an assessment of its performance. On the Institution's fiftieth
anniversary Vannevar Bush prepared such an analysis, together with a
statement of operating principles that continue to give guidance to our
efforts. On the sixtieth anniversary Caryl Haskins wrote a superb essay
detailing past accomplishments and putting the current activities of the
Institution in the context of the times. This seventy-fifth anniversary is
the occasion for another long-term view.

The Institution has modest financial resources and a total professional
staff that is tiny in comparison to that of many universities. Its accomplish-
ments, however, have been substantial. Hundreds of postdoctoral fellows
have received training at our departments, and many of them have gone
on to make outstanding contributions in their fields. Carnegie staff have
produced thousands of scholarly works including more than 800 books
published by the Institution and countless articles in professional journals.

Judging the significance of scholarly endeavors is difficult, and only a
few experts can do so with assurance. Some of our activities, however, have
led to important practical applications whose value is more easily assessed.

In addition, there are the very important contributions that the Institu-
tion made to the national defense in two world wars by furnishing leader-
ship in the mobilization of scientists well ahead of the declaration of w^ar.
Staff members also made substantial contributions in the application of
science to the war efforts.

The well-documented history of the Carnegie Institution reveals an
enormous number of activities and contributions. Only part of these could
be cited. The criterion for selection was that an activity or accomplishment
made a difference. That is, had the Institution not conducted its programs,
the history of science, technology, or the country would have been different.

As might be expected most of the examples of Carnegie achievements
are drawn from the past. Often decades must pass before the significance
of a program is fully evident. However, staff of our departments are
pioneering scientifically and practically in new directions. Moreover, the
policies of the Institution are especially effective in fostering creativity.
Discussion of these matters follows treatment of outstanding accomplish-
ments.

3

4 CARNEGIE INSTITUTION

LEADERSHIP IN WORLD WAR II

The many books about applications of science during World War II
make clear tlie importance of Vannevar Bush, then President of the
Carnegie Institution. He was instrumental in the creation of the National
Defense Research Connnittee. and served as its Chairman. Later he directed
the Office of Scientific Research and Development. His office, and the
nerve center of research and development for national defense, was the
Carnegie administration building at 1530 P Street in Washington.

By training and profession Bush was an electrical engineer. For most
of his career he had been associated with the Massachusetts Institute of
Technology, where he served as Vice President and Dean of Engineering
from 1932 to 1938. In 1938 he was elected President of the Carnegie
Institution of Washington. There is no record of the considerations that
went into the selection of Bush, nor of the factors that caused him to
accept the office. But we can assume that Bush and the Trustees were
following the events unfolding in Europe. Hitler was on the march: He
had already moved into the Rhineland, swallowed Austria, and rendered
Czechoslovakia defenseless. Whatever ambitions Bush might have had for
leadership in the coming conflict, one thing is clear. As second in command
at M.I.T. there was little chance that he would have a national role to play.

But as president of one of the nation's leading scientific organizations,
with headquarters in Washington, his status was much changed. Bush
already had some contacts in Washington as a member of the National
Advisory Committee for Aeronautics, and soon after becoming President
of Carnegie Institution, he was named Chairman of the N.A.C.A. This
brought him into even closer contact with official Washington, including
Congressional committees. One of the contacts that automatically followed
his move to Carnegie was with Frederic Delano, who served as Secretary
of the Board of Trustees from 1930 to 1948. Mr. Delano was an uncle of
Franklin Delano Roosevelt, and the younger man often counseled with
his relative.

In his book Pieces of the Action Bush mentioned nothing about his
relationship with Delano or his thoughts on the move to Washington. His
story begins in 1940 during the so-called phony war. At that time, many
people thought the Maginot Line in France would hold back the Germans
anrl that the I'nited States could avoid involvement. But many scientists
and engineers, including Bush, were dubious. The lack of preparedness in
this country made them uneasy. They were especially troubled that the
armerl forces were doing so poorly in applying science to military hardware.
Aside from improvements in aircraft and the Naval Research Laboratory's
work in developing radar, little was being done.

On the civilian side, there were the National Academy of Sciences,

REPORT OF THE PRESIDENT O

which had been created during the Civil War to advise the government,
and the National Research Council, an arm of the NAS established for
the same purpose during World War I. Both organizations had the same
weakness: Their charters specified that they were to advise the govern-
ment only when asked.

The atmosphere in Washington changed drastically in May and early
June 1940 as the German armies overran France and the Lowlands.
England was in peril, and the resources of Western Europe were available
to Hitler.

Bush was ready with a plan for a National Defense Research Com-
mittee. He had already established a good relationship with Roosevelt's
closest aide, Harry Hopkins, and so was able to bring his plan to President
Roosevelt under favorable sponsorship. It was set forth in four paragraphs
on one sheet of paper, and F.D.R. initialed it during a ten-minute meeting
with Bush.

In the next few days Bush talked the matter over in greater detail with
Harry Hopkins and submitted to the President a list of proposed members
of the N.D.R.C. These included F. B. Jewett, President of the National
Academy of Sciences, and Conway P. Coe, Commissioner of Patents, both
to serve ex officio. Other members were J. B. Conant, President of Harvard
University; Karl Compton, President of Massachusetts Institute of Tech-
nology; and Richard C. Tolman, Dean of the Graduate School, California
Institute of Technology. Later Roger Adams, head of the Department of
Chemistry, University of Illinois, was added as well as representatives of
the Army and Navy.

In Pieces of the Action Bush made what was for him an unusually
blunt statement :

There were those who protested that the action of setting up
N.D.R.C. was an end run, a grab by which a small company of
scientists and engineers, acting outside established channels, got
hold of the authority and money for the program of developing
new weapons. That, in fact, is exactly what it was. Moreover,
it was the only way in which a broad program could be launched
rapidly and on an adequate scale. To operate through established
channels would have involved delays — and the hazard that inde-
pendence might have been lost, that independence which was the
central feature of the organization's success. The one thing that
made launching it at all possible was the realization by the
President that it was needed.

Immediately after obtaining a charter, the N.D.R.C. set up an effective
organization, and soon top scientists were responding eagerly to invitations
to participate.

There was never any question who was Chairman of N.D.R.C. or, sub-

6 CARNEGIE INSTITUTION

sequent ly. who was Director of the Office of Scientific Research and De-
\elopnient. A chain of command and a pyramid structure \vere maintained
at all times. But tlie chain carried connnunications in both directions. A
good idea could easily move from below to the top.

No research and development was carried out at headquarters. But there
were an enormous number of tasks to be performed in Washington if
defense research was to be conducted smoothly and effectively. Besides, a
new device was worthless unless it was manufactured, deployed, and used
properly. Bush was superb in all aspects of expediting.

A most impressive feature of Bush's performance was the surefooted way
he operated in what is to many the Washington jungle. He enjoyed in-
creasingly cordial relations at the White House, and Roosevelt came to
depend on him as his science adviser. Bush won the good will of the rele-
vant committees of Congress, and he managed to evade problems created
by bureaucrats. Establishing good w^orking relations with the Armed
Forces took some doing. There was suspicion at mere civilians dabbling in
matters considered to be in the province of the military. There was also a
natural concern about secrecy. Later these attitudes changed as a result of
achievements and the demonstrated ability to keep secrets.

In technical matters. Bush made decisions quickly. He sought the best
counsel, but even w^hen his advisers were uncertain, he was decisive. In
his directing of scientific efforts for national defense. Bush brought to the
situation a remarkable set of talents: good judgment, organizational know-
how, political adroitness in the best sense of the words, decisiveness, and
integrity. The Carnegie family is proud of his accomplishments and pleased
that the presidency provided his base of operations.

Proximity Fuze

In terms of the effect on the course of combat, the VT (variable time)
proximity fuze was probably the most important technical improvement
in weaponry to come out of World War II. It was essentially a radio set in
an artillery shell. When the shell came close to a target, the radio circuit
triggered the detonation. Before the war no such device existed, but by
the end of the war tens of millions of the fuzes had been assembled at a
cost of many billions of dollars. Artillery using such shells was five to ten
times more effective in knocking down enemy aircraft, for example, than
artillery using conventional contact detonators. The fuze was extremely
useful to the Navy in the Pacific in destroying enemy bombers and kami-
kazes. It was crucial against the V-1 buzz bombs: On the last day of those
attacks, 100 of 104 buzz bombs failed to reach their targets. The VT fuze
was used in ground combat for the first time in December 1944, when it

REPORT OF THE PRESIDENT /

helped turn back the Germans' Ardennes offensive (the ^'Battle of the
Bulge"). The enemy was astounded by the effectiveness of our artillery fire.

Many individuals had independently thought of making a proximity
fuze. But a few moments' consideration of the practical problems stopped
most of them. When a shell is fired from a gun, it undergoes an accelera-
tion 20,000 times that of gravity. In addition, the rifling of the barrel
imparts a rapid spin to the shell with correspondingly great centrifugal
forces. Vacuum tubes then commercially available could not withstand
such forces. If the tubes were strengthened, vibration of the tube elements
during flight could lead to premature firing. A further constraint was that
the fuze had to be small enough to fit inside an artillery shell.

One Sunday in August 1940, Merle Tuve burst into the laboratory of
a colleague at the Carnegie Department of Terrestrial Magnetism (DTM).
He had been listening to radio reports of a massive air raid on London,
and he was greatly disturbed by the destruction and the civilian casualties.
He spoke with determination of the necessity to develop better antiaircraft
defense.

On August 17, 1940, Tuve received the directive he needed from the
National Defense Research Committee to develop ^^influence" fuzes. At
that time Tuve, the Chief, had at his disposal four Carnegie ^'Indians" —
Lawrence R. Hafstad, Richard B. Roberts, George K. Green, and Robert
Myers. They were energetic, innovative physicists, but their experience
with ordnance amounted to a few weeks in summer R.O.T.C. camp.

But by the evening of the day Tuve got his go-ahead, they had already
shown that some vacuum tubes could withstand 5000 g, and in the next
few days their results were even more encouraging. They put a tube inside
a lead ball and dropped it from the top of a building at DTM onto a
concrete pad. From the indentation of the lead, they were able to show
that the tube had withstood a force of 20,000 g. They realized, though,
that an acceleration of short duration was not equivalent to the longer
effect in a gun, so they built their own cannon for more realistic testing.

In the months that followed, Tuve expanded the scientific and engineer-
ing staff at DTM to 100. He also negotiated contracts with outside com-
panies for development of rugged components. At the same time he w^as
already looking ahead to the use of proximity fuzes in combat. For example,
he was concerned that gun crews should not be exposed to the hazard of
premature explosion, so he saw to it that the fuzes were designed with
many safety features.

During the succeeding phases, encouraging developments were followed
by a continuing series of almost crippling discouragements. It seemed
that everything that could go wrong went wrong. But the effort to over-
come unexpected problems was relentless. Failures were analyzed and the
necessary countermeasures devised.

8 CARNEGIE INSTITUTION

By January 1942 a model of the VT fuze had been successfully tested
by the Navy, and the fuze was in pilot-scale production, with full-scale
production ordered. At that point the cost of development to the govern-
ment was only SI million. In March the proximity fuze project was trans-
ferred from DTM to the new Applied Physics Laboratory of The Johns
Hopkins University, where development continued on an expanded scale
with Tuve still in charge. ]\Iany other models of the fuze were devised
there, and large-scale production was monitored.

Tuve was a tough taskmaster who did not allow his organization to
become complacent from success. There was much work to do in improving
reliability of the fuzes and in tailoring them to be more highly effective
against specific types of targets. Something of the philosophy with which
the organization operated was encapsulated in the following quotation
from Xew Wcapo?is for Air Warfare, edited by Joseph C. Boyce:

Nothing can iUiistrate the spirit with which Tuve infused his
workers hetter than the following representative set of ^'Section
T Verbal Rules," or unwritten informal ^'operating rules." The
elastic organization of the Applied Physics Laboratory allowed
full scope for the operation of these rules. Under this code of
operation it was surprising to note to what degree the initiative
and personal responsibility of staff members expanded. . . .

Operating Rules

1. "1 don't want any d — n fool in this laboratory to save
money, I only want him to save time.

2. "We don't want the best unit, we want the first one.

3. ''There are no private wires from God Almighty in the lab
that I know about — certainly none in my office.

4. "The primary duties of any supervisor are initiative and
forethought, he is supposed to make his team do the work.

5. "Any function or area of a total job which can be described
and manned should be assigned. Articulate your work.

6. "The trouble is always at the top.

7. "The Navy says so! Who is the Navy? That was only the
opinion of the man you were talking to.

8. "A good short paper in your hand at the right instant and
place is a marvelous hatchet for getting action. ^Red tape' is a
tool; use it, but use discrimination in your paper work.

9. "Responsibility and authority always have the same bound-
aries; this is axiomatic.

10. "Our moral responsibility goes all the way to the final
battle use of this unit; its failure there is our failure regardless
of who is technically responsible for the cause of failure. It is our
job to achieve the end result.

REPORT OF THE PRESIDENT 9

11. ''Run your bets in parallel, not in series. This is a war
program, not a scientific program.

12. 'The final result is the only thing that counts, and the only
criterion is, does it work then.

13. "Shoot at an 80% job, just a passing grade, we can't afford
perfection. Don't try for an A, in a war a D is necessary and
enough but an F is fatal. Don't forget that the best job in the
world is a total failure if it is too late."

The consequences of this pragmatic philosophy were displayed in con-
nection with the German buzz bomb. Tuve learned that the enemy might
be developing such a device. Long before the British were convinced of
the need for ground-based anti-buzz bomb devices, the appropriate fuze
had been developed at the Applied Physics Laboratory, production had
begun, and large quantities of fuzes had been sent to England. The British
attempted to cope with the buzz bomb using their aircraft. This effort was
only partially effective. When the task was assigned to artillery located
near the coast of England and equipped with the proximity fuze, the buzz
bomb attack was stopped.

LEADERSHIP IN WORLD WAR I

When Vannevar Bush took the initiative in mobilizing scientific effort
for national defense during a world war, he was not the first Carnegie
person to do so. George Ellery Hale, then Director of the Mount Wilson
Observatory, had played a similar role in World War I.

In July 1915, shortly after the sinking of the Lusitania, he proposed
to the President of the National Academy of Sciences that the services
of the Academy be offered to President Wilson for the purpose of organiz-
ing scientific research agencies to foster national preparedness. This pro-
posal was not acted on immediately. But Hale was Foreign Secretary of
the Academy and thus a member of its Council, and another submarine
attack created an opportunity for him to try again. This time he brought
the matter up himself, and in April 1916 the Council adopted the follow-
ing resolution :

Resolved, That the President of the Academy be requested to
inform the President of the United States that in the event of a
break in diplomatic relations with any other country the Academy
desires to place itself at the disposal of the Government for any
service within its scope.

President Wilson accepted this offer and asked that the Academy coor-
dinate the country's scientific resources and secure the cooperation of all

10 CARNEGIE INSTITUTION

agencies, governmental, academic, and industrial, in which research facili-
ties were available.

Hale was named chairman of the organizing committee whose activities
led to the formation of the National Research Council on September 20,

1916. Hale served as Chairman of the NRC until April 30, 1919. Under
his leadership there was substantial mobilization of scientific effort.

Hale did not confine his attention to the United States. In the spring
of 19 IS. he submitted to the Council of the National Academy a ^'Sug-
gestion for the International Organization of Science and Research," partly
to secure closer cooperation among the scientific men of the Allies in the
solution of war problems and partly to build a framework for international
cooperation in research after the war. This suggestion was adopted by the
Council and Hale, as chairman of the American delegation, presented it
at a conference of Academies called by the Royal Society in London in
October 1918. Hale's initiative led directly to the creation of international
scientific unions and to a superstructure to coordinate them — the Inter-
national Council of Scientific Unions, which is today a vital force for
bringing about international scientific cooperation.

Optical Glass

At the beginning of World War I, Jena in Prussia was the world's center
for the manufacture of high grade optical glass. In consequence, the
Germans enjoyed superiority in the various devices that required such
glass, including scientific instruments and military optics. The United
States manufactured ordinary glass, but the product was inferior for optical
applications. When the United States entered the war, a serious shortage
of optical glass was felt at a time of enormous demand for new ordnance.

Scientists at our Geophysical Laboratory had no experience in the com-
mercial production of glass. But their laboratory experience with silicate
melts and especially their understanding of phase equilibria in silicate
systems were unique, and that knowledge proved to be decisive. Leason H.
Adams and E. D. Williamson quickly outlined a faster, superior method
of annealing. Frederick E. Wright devised a highly sensitive technique of
quality control of the finished glass.

In April 1917, Arthur L. Day, Director of the Geophysical Laboratory,
was asked to guide research to aid the country's optical glass production.
Fred Wright went to the Bausch & Lomb plant in Rochester, New York,
anr] applied his knowledge to improving production. At the same time, he
became familiar with the techniques of commercial glassmaking. By June

1917, Wright had been put in charge of production of optical glass at
Bausch (fe Lomb and was turning out a high quality product. The progress
there was so impressive and the demand for glass so great that the Geo-

REPORT OF THE PRESIDENT 11

physical Laboratory was invited to take charge of production at two other
major plants.

The United States had entered the war in 1917 importing all its optical
glass. By late in the war, it was a net exporter of optical glass. Of the
675,000 pounds of glass produced for war purposes, 95% was made under
the direction of the Geophysical Laboratory, with 10 men in the field and
13 at work at the Laboratory.

PEACETIME ACTIVITIES

During major wars members of the Institution showed their flexibility
in tackling highly practical tasks. Their efforts involved high priorities,
short deadlines, and management decisions affecting many thousands of
lives.

During peacetime, the tempo and outlook of staff members are com-
pletely different. Instead of secrecy, the policy is complete openness. In-
stead of a highly directed effort, the guiding principle is maximum indi-
vidual initiative. Instead of immediate objectives, the goals are long range.
Instead of emphasis on quick practical applications, there is a search for
understanding. But at Carnegie the ivory towers have windows, and the
staff are aware of society's needs.

Thus it should not be surprising that some of the fundamental research
conducted by the Institution has resulted in important practical ap-
plications. These include hybrid corn, radar, geophysical prospecting
equipment, improved ceramics, refractories, glass, and cement. These
achievements alone abundantly justify the existence of the Institution.
Measured in dollars, the cumulative benefit to society of development of
hybrid corn has been more than $100 billion. In contrast, the total endow-
ment of Carnegie Institution was $32 million and the cumulative expendi-
tures have been about $180 million.

But the major contributions of the Institution are not measurable in
financial terms. They are parts of a grand edifice of knowledge in the
design of which the Institution has had a significant role.

Department of Terrestrial Magnetism

During the first half of its existence, beginning in 1904, the Department
of Terrestrial Magnetism maintained a program consonant with its name.
But in the last three decades the activities have been much broader than
the name, and the Department has been a leader and made major contri-
butions in many fields. Its role in creating and fostering the discipline of
geophysics was crucial.

1-3 CARNEGIEINSTITUTION

When Carnegie's Department of Terrestrial Magnetism was established,
the earth's magnetic and electric fields had been explored in only a few
areas. Little was known about the oceans, covering three quarters of the
area of the globe, or about the vast polar regions. To fill in these gaps in
oiu' knowledge of the earth, the Department launched a major program
oi magnetic observation. From 1909 until her destruction by fire in 1929,
the nonmagnetic brig Carnegie made seven long cruises, covering all the
oceans and reaching the boundaries of the north and south polar-ice areas.

In those days, knowledge of the earth's magnetic field was particularly
useful for ships at sea. During World War I, the staff helped develop a
magnetic navigational device for aircraft, and they were a little ahead of
the times in developing the magnetic mine.

From about 1920 to 1940. the Department was the world's leading center
for geophysics. One of the most far-reaching events of that period was the
invention of radar and its use in studies of the ionosphere. Commenting on
this development in Carnegie Year Book 51, Vannevar Bush wrote, "Radar,
in its essence, is the method of locating objects in space by propagating a
beam of short pulses of electromagnetic energy and measuring the time
between the pulse and its echo at the sending station. The pulsation is an
essential element of the process. On July 10, 1925, Merle A. Tuve and
Gregory Breit, then both of the Department of Terrestrial Magnetism,
sent out pulses from a Navy transmitter they had modified for the purpose
and observed echoes from the ionosphere. During the next few years, as
they were studying the ionosphere, they were troubled by echoes coming
from airplanes, which interfered with their measurements. These experi-
ments introduced the first effective electrical method of surveying the
ionosphere and the first observation of the echoes of pulses from airplanes."
Later this radar method was developed further by Lloyd V. Berkner and
others to permit measurement of variations of the electron density and
heights of ionized layers of the ionosphere. Such changes in the ionosphere
are closely related to radio communication. Suitable practical and auto-
matic apparatus was developed to send radio pulses yielding echoes from
the ionosphere which were continuously recorded at a wide range of
frequencies. These measurements provided a record of events throughout
the range of about 100 to 300 km and higher within the ionosphere, for
study in connection with geomagnetic and other phenomena observed at
ground level.

Based in part on the studies of oceanic electrical measurements con-
ducted aboard the Carnegie, staff at DTM discovered that the negative
electric charge over the earth's surface as a whole varied periodically by
about 30% of its average magnitude every 24 hours, independently of
local time. This finding led two Staff Members, George R. Wait and Oliver
H. Gish, to investigate the role of thunderstorms. In 1947 and 1948 Wait

REPORT OF THE PRESIDENT 13

and Gish, then close to retirement age, made 65 traverses over the center
of thunderstorms at altitudes of up to 48,000 feet. They found that in
some storms a substantial current flowed in a direction opposite that noted
in fair weather. From these observations they were able to prove that
thunderstorms are the source of the current maintaining the earth's nega-
tive charge, which would otherwise be reduced to a fraction of its observed
value in a matter of minutes.

Gish, along with William J. Rooney, was also responsible for new experi-
mental approaches to the study of natural and induced earth currents.
They found that the time variations in earth-current potentials are pro-
duced by variations in geomagnetic fields of external origin. One practical
result of their studies was a technique of measuring earth resistivity which
is still widely used in geophysical prospecting.

To the staff of DTM, physics has never been just a matter for laboratory
study. Like the long, dangerous voyages on the Carnegie and the flights
of Wait and Gish over electrical storms, DTM's worldwide observatory
program has been a means of recording physical phenomena right at the
source. For example, in the 1930s seismographs were installed at Huancayo,
Peru. And Carnegie's Cosmic Ray Committee sponsored a network of
observatories in Maryland (USA), Peru, New Zealand, Mexico, Greenland,
and Puerto Rico. Staff Member Scott Forbush used data from these sta-
tions to show that the sun emits cosmic rays.

Some of the other terrestrial-solar phenomena that are better under-
stood because of the Department's observatory program are: the deter-
mination that the daily and lunar geomagnetic variations are caused by
the ultraviolet radiation from the sun and the motions of the upper
atmosphere in the presence of the earth's magnetic field; closer linkage of
specific (solar flare) and average solar conditions with variations in geo-
magnetism, the ionosphere, aurora, and cosmic rays; and the description
of worldwide geomagnetic variations of various periods at different times
in the sunspot cycle and their possible electric current systems in the
atmosphere. The results of the observatory program were also useful in
practical applications such as the improvement of long-distance radio
transmission.

Nuclear Physics

In the 1930s and shortly thereafter, DTM was a leading center for
nuclear physics. The study of the atomic nuclear reactions was then just
beginning in this country. Merle Tuve pioneered in the construction and
use of Van de Graaf generators for the observation of nuclear forces and
reactions. The paper by Tuve, Lawrence R. Hafstad, and Norman Heyden-
burg on proton-proton scattering was widely hailed. The work established

IJf. CARNEGIE INSTITUTION

the nature of forces existing when two charged bodies come close together.
This was in effect the first measurement of the diameter of a proton
(10~^^ cm). Tuve also served as a catalyst in the interaction of nuclear
theoretical physicists. Superb conferences on this topic were held each
year: they included leaders like Bohr, Fermi, Wigner, Bethe, Teller, and
Gamow. Out of one of those conferences came Bethe's first explanation
of nuclear reactions in the sun. At another conference, in January 1939,
Bohr first disclosed to an American group the discovery of uranium fission.
That night the discovery was confirmed at DTM. Shortly after that,
Richard Roberts and Robert C. Meyer discovered delayed emission of
neutrons during uranium fission. This delayed emission is crucial to the
control of present-day reactors.

Early in 1940 P. H. Abelson arrived independently at the hypothesis
that element 93 (neptunium) was being formed during irradiation of
uranium by neutrons. He devised a chemical method for testing this
hypothesis and proved it to be correct in joint research with Edwin Mc-
Millan at Berkeley. Proof of the existence and properties of neptunium
was obtained in five days of effort. He then developed a liquid thermal
diffusion method for separating uranium isotopes. Work on the process
was later fostered by the Navy and the method was ultimately used in
the ^lanhattan Project to speed the production of the first substantial
amounts of U^^^.

Biophysics

The time immediately following World War II was one of reevaluation
and new beginnings. Tuve, Roberts and Abelson agreed that to participate
at the forefront of nuclear physics would involve building huge and costly
accelerators or working on secret projects such as new weapons systems.
But that kind of research would have been incompatible with a peacetime
Carnegie Institution. They wanted to engage in research more likely to
leave a constructive residue. The group decided that biophysics held the
opportunities they were looking for. Their reasoning was that physicists,
with their penchant for analysis and separation of variables and their
ability to devise new instrumentation, might be able to crack some of the
complex puzzles in the realm of living matter.

One of their first projects was a series of experiments using radioactive
tracers that permitted insights into the dynamic character of biological
processes. By 1951 the Biophysics Section had crystallized. There were
four physicists — Roy J. Britten, Dean B. Cowie, Roberts, and Abelson —
and a biologist, Ellis T. Bolton, whose background included service as a
radar officer in the Marine Corps during World War II.

REPORT OF THE PRESIDENT 15

The group concentrated on studies with the bacterium Escherichia coli.
That organism had been chosen because of its supposed simplicity. Actu-
ally, within a volume of a fraction of a cubic micron, thousands of
coordinated chemical reactions go on in E. coli, and thousands of enzymes
function in harmony. Given proper nutrients and conditions, the organisms
are highly dependable. They are also very efficient.

Using radioactive tracers, the Biophysics Section explored the marvelous
capabilities of that very competent chemical engineer, E. coli. For example,
it was found that when a chemical that E. coli would ordinarily synthesize
and use is added to the medium, in many cases the bacteria stop mak-
ing the chemical and obtain it from the medium. And if a substance
is added that is a chemical intermediate in the synthesis of a final product,
the bacteria may stop synthesis of the intermediate and pick it up instead
from the medium. This discovery enabled the group to work out in detail
the synthetic pathways. Supplementary studies with other microorganisms
showed that they use many of the same kind of synthetic pathways found
in E. coli. These studies led to numerous journal articles and were brought
together in a book published in 1955, Studies of Biosynthesis in Escherichia
coli, which won acclaim from microbiologists around the world.

There were also benefits from the activities of the Biophysics Section
not measured in the scientific literature. Members of the group shared
in an exciting learning experience in which each taught the others. At the
time, they were in their late thirties, but intellectually they were like
youngsters in university. They experienced a thrilling rejuvenation.

Abelson left the Biophysics Section and DTM in 1953 to become Di-
rector of the Geophysical Laboratory. There he used some of the knowledge
and techniques he had acquired in a new venture in organic geochemistry.
The Biophysics Section continued its creative ways. In the next 20 years it
produced about 200 professional papers and two additional books. Roberts
listed 18 of the Section's achievements in an historical summary which
appeared in Year Book 7^.

Two major areas of the later studies stand out. One had to do w^ith
macromolecules. A second involved the structure of chromosomes.

In a series of imaginative experiments, the group was able to show for
the first time that the machinery of protein synthesis consisted of ribo-
somes. Other studies of ribosomes led to four papers describing the kinetics
of synthesis of ribosomes including both nucleic acid and the protein
components. The group also demonstrated that in newly formed bacterial
RNA, one third is messenger RNA and the rest, ribosomal precursor.

A useful technique for studying RNA was to immobilize single-stranded
DNA on an agar column. Complementary RNA could then be immobilized
on the bound DNA. Complementary DNA could also be held up. This
technique, invented by Bolton and Brian J. McCarthy, led to another

16 CARNEGIE INSTITUTION

column procedure in which the mineral hydroxyapatite formed the column
bed. Double-stranded DNA is bound to the column while single-stranded
DXA passes through. These techniques led to studies of the relatedness of
DXAs from various organisms. Later, they also made possible an important
finding about the structure of DNAs from all animals.

When chromosomes are heated in a solution to about 85°C, the double
strands dissociate into single strands. If the solution is cooled to about
GO^'C. the strands find their appropriate complementary mates and re-
associate rather quickly. If the chromosomes are cleaved into smaller pieces
and the heating and partial cooling procedure is followed again, the pieces
reassociate. If unrelated DNA is introduced, it does not interfere with the
reassociation.

Using these techniques, Britten made discoveries that have been de-
scribed as the most important advances in genetics of the past 15 years.
He found that chromosomes of an animal contain many repeated identical
structures. For several years, most of the work of the Biophysics Section
was devoted to exploring the ramifications of this finding.

Among the discoveries were: (1) Repeated DNA sequences were noted
in all species tested in all the various phyla above the fungi. (2) From
20'7r to SO^r of the total nuclear DNA was repeated DNA. (3) In a given
genome there were many families each having from 50 to 20,000,000
related sequences. (4) The degree of relatedness within families ranged
from bare detectability to nearly perfect identity. (5) The repeated DNA
sequences were scattered throughout the length of the genome. (6) Pat-
terns of frequency of repeated DNA vary widely even among vertebrates.
(7) RXA complementary to some repeated DNA sequences has been ob-
served in every cell type examined. (8) Different sets of repeated sequences
are transcribed in different tissues and stages of development.

This significant work could be undertaken because of the unique struc-
ture of the Carnegie Institution, which encourages to an unusual degree
the taking of creative scientific risks. In few if any other places in America
could a handful of scientists previously untrained in a discipline be
nurtured in it to a point of such high achievement.

Crustal measurements. In 1946 Tuve and his co-workers undertook the
first continuing program of deep crustal measurements with instruments
adequate to the task. At that time the ideas of R. A. Daly, H. Jeffreys,
and B. Gutenberg concerning a simple layered structure of the crust were
very influential, and questioning those ideas was not fashionable among
seismologists.

For the next seven years the DTM group had this field nearly to
itself. New instruments were developed which improved the sensitivity
to the subangstrom level, and new techniques, such as subcritical reflec-
tions, were used to identify the base of the crust. Tuve's group found that

REPORT OF THE PRESIDENT 17

the evidence for a layered crust was more of a mathematical convenience
than a geophysical reality, and that the thickness of the crust was highly
variable and not in accordance with previous notions of isostatic com-
pensation. Various seismic arrivals which had been interpreted as indicating
crustal layering were found to be only locally coherent and of no funda-
mental significance, and p-wave velocities in the uppermost mantle were
shown to be variable. All this went against prevalent geophysical theory,
but gradually the power of this method and the significance of these results
were recognized. These pioneering studies gave seismologists the tools they
needed ten years later, when nuclear test detection became an issue. The
seismic community took the opportunity to increase our knowledge of
the earth while carrying out a practical political task.

In contrast, the need for a practical method of measuring geological
time was generally recognized in the late 1940s. But the direction opened
so brilliantly by University of Minnesota physicist A. 0. C. Nier during
the 1930s was not favored, on the grounds that the required instrumenta-
tion and chemical techniques were beyond those feasible in a geological
laboratory.

The DTM group, led by L. Thomas Aldrich, with collaborators at the
Geophysical Laboratory, took up where Nier left off and made the measure-
ments needed to really understand what was going on with isotopes in
rocks. The currently accepted decay constants for potassium and rubidium
were found to be seriously in error. The conditions under which natural
systems were chemically open were identified, and the basic distribution
of rocks as a function of age was established for several continents.

By about 1958, it was recognized that the Carnegie approach was su-
perior, and the field grew rapidly during the next decade. By 1969 when
the lunar samples were distributed and the Allende meteorite fell, the
techniques of K-Ar, Rb-Sr, and U-Pb dating had been developed into
very accurate and sophisticated tools suitable for the job of measuring
the small but significant isotope differences found in these materials. The
interpretation of these results has had far-reaching consequences in our
present concept of the development of the solar system.

Image tubes for astronomy. In the mid-1950s President Bush established
a major program within the Institution to develop electro-optical devices
that could increase the sensitivity of astronomical telescopes by a factor of
10 or more. This cooperative program, involving many institutions and
industrial laboratories, was organized at DTM under the chairmanship of
Tuve. The effort was directed toward developing and demonstrating elec-
tronic image intensifiers that were practical for routine observational use.

From the beginning, the goal of the Carnegie Image Tube Committee
was to develop a device that would make smaller telescopes as effective
as larger telescopes with conventional detectors. In the early 1960s such

IS CARNEGIE INSTITUTION

a system was demonstrated. From 1963 to 1970 the Carnegie Committee,
with National Science Foundation funding, built and distributed 34 highly
successful image intensifier systems to observatories in this country and
abroad. For many observatories, these ^'Carnegie Image Tubes" have be-
come the major observing tool and have made possible many varied
astronomical observing programs that otherwise could not have been
undertaken.

Image tubes also increased the power of the largest telescopes. Success
with image tubes stimulated the development of other electronic devices for
use at the observing end of telescopes. This trend is continuing, and it
represents a revolutionary development in astronomy.

Geophysical Laboratory

In many ways the Geophysical Laboratory occupies a unique position in
a field where science and technology overlap. It has made major contri-
butions in both pure and applied research. The Geophysical Laboratory
scientists have contributed to physical chemistry and geology, and as a
group they have served as leading interpreters and advocates of the appli-
cation of the exact sciences, including physics and chemistry, to the solu-
tion of problems in the earth sciences. Examination of five of the principal
advanced textbooks on petrology reveals that about one fourth of the
literature cited covers work that was performed at the Geophysical Labora-
tory. The Laboratory has also made important contributions through its
work in mineralogy and volcanology. The establishment of the scale for
high temperatures, the phase equilibrium studies, and other physico-
chemical investigations have earned the Laboratory a niche of its own in
the physical sciences. Today the Laboratory is one of the leading centers
of investigation of chemical phenomena at very high pressures.

Experimental results obtained at the Geophysical Laboratory have had
important applications in many industries. The basis for modern manu-
facturing procedures in the Portland cement industry, for instance, is a
1010 paper by George A. Rankin and Fred Wright on the melting be-
havior of mixtures of lime, alumina, and silica ('The ternary system CaO-
ALO.'.-SiOii"). Their article has been called ''the bible of the Portland
cement industry" because it answered scientifically the long-standing
questions about cement clinker composition and the setting and hardening
characteristics of Portland cement. Another technological application of
the Laboratory's researches is in glass manufacture. A paper by G. W.
Morey and N. L. Bowen showed for the first time which of the ternary
melts could be readily quenched to glass. This work standardized and
improved the manufacture of glass worldwide. With J. W. Greig, Bowen
also wrote "The system Al20.',-Si02," which showed that the place to look

REPORT OF THE PRESIDENT 19

for improved refractories was not in the 1:1 molecular ratio of the two
oxides, as had been supposed, but in the 3:2 ratio of the new mineral
mullite.

Major programs of the last two decades have included extensive studies
of the behavior of silicates at high temperatures and pressures in the
presence of H2O and CO2. These studies duplicate conditions deep in the
earth. Another major trend has been the development of a series of equip-
ment capable of unprecedented high static pressures. Another notable
program has been a successful pioneering effort in organic geochemistry.

Studies in igneous petrology have been a major activity of the Geo-
physical Laboratory. This emphasis was established early with the devel-
opment of the quenching method for studying the melting relations of
silicate materials, and the Laboratory continued on the forefront of petrol-
ogy on the basis of this type of research for more than half a century. Its
success in this field is often attributed to the highly perceptive work of one
staff member in particular, Bo wen, but it was also the contributions of
many other scientists at the Laboratory that produced the high reputation.
These men demonstrated many of the important principles and processes
of igneous petrogenesis, such as (1) the role of fractional crystallization
in producing a diversity of igneous rocks, (2) the reaction principle
whereby one mineral is resorbed in a cooling melt while another crystal-
lizes, (3) the nature of the processes that occur when foreign material is
assimilated by magmas, and (4) the nature and role of liquid immiscibility
in magmas. Pioneering studies were made on such important topics as
(5) the solubility of water in magmas and diffusion in silicate melts, and
at the same time there was produced a tremendous body of (6) basic data
on equilibrium relations between crystals and melt in a wide variety of
silicate systems approximating igneous rocks in composition.

To complement the experimental studies, field investigations of the
igneous rocks themselves were frequently carried out. This work did much
to guide the direction of the experimental research and included classic
studies on particular areas, such as the Hawaiian and Aleutian Islands,
and on specific rock suites such as ultramafic rocks and anorthosites.

In a survey article published in 1974, the late W. W. Rubey, one of
the most distinguished geologists of this century, evaluated some of the
great developments of the 50 years of the earth sciences. Referring to
contributions of the Geophysical Laboratory, he wrote:

In petrology, major new developments have profoundly modi-
fied the course of research on problems of igneous, metamorphic,
and sedimentary rocks. Perhaps the most far-reaching develop-
ment, in its consequences, has been the application of the prin-
ciples of physical chemistry to the study of igneous, and later,
metamorphic rocks. . . . The basic principle guiding the program

'20 CARNEGIE INSTITUTION

at this laboratory has been that all experimental work is done
on materials of known purity and under accurately known
conditions. . . .

Bowen's differentiation theory has had tremendous influence
in this country an(i abroad, although questions still remain about
how universally it may be applied. However, it is })robably fair
to state that the science of petrology has been made over by the
results of exjierimental work at the Geophysical Laboratory and
by Bowen's theory. Perhai)s an even greater contribution of the
Laboratory is one that extends beyond petrology into many other
fields of science. This is the demonstration that j^'oblems of
seemingly hopeless complexity may be divided into simj^ler, more
tractable parts and brought into the laboratory for analysis and
solution.

The Geophysical Laboratory has an unexcelled record in the develop-
ment of equipment that opens new^ frontiers for experimentation. In recent
years there have been innovations in apparatus for amino acid analysis,
for the electron microprobe, for high-temperature and high-pressure crys-
tallography and for dating rocks.

A major continuing contribution has been in the development of equip-
ment for general high-pressure, high-temperature research.

The extent to which knowledge of the earth's interior can be gained
depends largely upon the capability of high-pressure apparatus used in
the laboratory to simulate the pressure-temperature conditions of the
earth. Equipment developed at the Laboratory has been adopted by most
of the major high-pressure laboratories all over the world. ^'One of the
striking features of the [Geophysical] Laboratory," wrote R. S. Bradley
at the beginning of his series of books High Pressure Physics and Chemistry,
''is the way in which visiting scientists, of whom the writer was one, are
encouraged to pick up the know-how; they also benefit greatly from an
injection of enthusiasm."

The record of development of high-pressure apparatus during the past
three decades has been impressive. The cold-seal pressure vessel, designed
by Tuttle, has been adopted worldwide as the most commonly used vessel
for hydrothermal studies. The single-stage, solid-media apparatus designed
by Boyd and England has been widely copied. It is estimated that about
200 of such single-stage units are now being operated in laboratories in
the United States and about 100 in Europe.

H. K. Mao joined the Laboratory in 1968 and initiated the extensive
development of a diamond-anvil high-pressure cell, a concept originated
by A. Van Valkenburg some years ago. A diamond-anvil cell recently de-
signed by Peter M. Bell and Mao is capable of maintaining 2000°C at
1500 kbar when heated by the beam of an Nd-doped YAG laser. The
sample between the diamond anvils can be observed at high pressure by

REPORT OF THE PRESIDENT 21

microscopy, infrared spectroscopy, x-ray diffraction, and Mossbauer spec-
troscopy. In addition, the electrical conductivity and optical absorption
of the sample can be measured at high pressures.

The new diamond-anvil cell extends greatly the range of static high
pressure in the previous apparatus and offers a powerful tool for studying
the behavior of materials at great depths in the earth. It gives access to
phenomena pertaining to about 85% of the earth by volume but about
which little was known owing to previous limitations of high-pressure
apparatus. Because of the versatility of the diamond-anvil cell, a broad
spectrum of physical and chemical properties of the lower mantle can now
be studied.

Research at the Geophysical Laboratory has in general related to the
petrology of igneous and metamorphic rocks. But the sedimentary rocks
which cover most of the surface of the earth also present important puzzles
and phenomena to investigate. The geochemistry associated with sedi-
mentary processes is greatly affected by the activities of living matter,
which can create a highly reducing environment and which produces vast
quantities of organic chemicals. The amounts that appear as petroleum,
natural gas, coal, and oil shale are only a tiny fraction of the organic matter
of the earth's crust.

Most of what is known about past life forms has been inferred from the
fossil record, which is satisfactory during the last 600 million years but
sparse and not very informative for 3000 million years before that. More-
over, while the hard parts of the fossils are important, the essence of life
resides in organic matter.

The organic matter that is formed each year is, for the most part, rela-
tively quickly destroyed, but some escapes destruction and persists for long
periods. What can studies of the organic constituents of rocks tell us about
ancient life? How do organic chemicals in rocks change with time as a
function of the conditions to which they are exposed?

A notable beginning to examination of such questions was made by
A. Treibs during the 1930s in his studies of porphyrins, but the work then
languished. In 1953, Abelson began work that stimulated new interest in
organic geochemistry. He discovered that many fossils contain amino acids.
The initial work was performed on shells of the clam Mercenaria mercenaria,
commonly known as the quahog. This mollusc also flourished 20 million
years ago and fossils from that period contain some of the amino
acids present in modern shells. It was apparent that the dense mineral
structure of the shell (aragonite) had protected at least some of the amino
acids from bacterial attack and that the chemicals remaining had possessed
sufficient chemical stability to last 20 million years. The absence of some
of the amino acids suggested that they lacked long-term stability. This
hypothesis was tested by conducting incubations of amino acids for ex-

1^2 CARNEGIE INSTITUTION

tended periods at moderately elevated temperatures. An excellent correla-
tion between laboratory tests and fossil data was fomid. The compounds
that remained in the fossils possessed the greatest stability in laboratory
tests. These findings made it evident that under favorable conditions
interesting biochemicals (biochemical fossils) could endure for hundreds
of millions of years. By the early 1960s the study of biochemical fossils
was being conducted in a number of laboratories and information derived
from such studies is finding application in petroleum exploration.

Many other pioneering studies related to organic geochemistry were
conckicted at the Laboratory, including epimerization and isomerization
of amino acids, use of changes of amino acids as a dating tool, mechanisms
of formation of humic acid and kerogen and the low temperature conver-
sion of kerogen to hydrocarbons.

Two lines of experimental work, one by T. C. Hoering and Abelson, and
the second by Hoering, produced strong evidence that organisms living
more than 3000 million years ago engaged in photosynthetic processes at
least in part similar to those employed today. The evidence came from
studies of carbon isotope fractionation that accompanies photosynthesis.
They measured isotopic fractionation in some present-day widely different
photosynthetic organisms, some thought to be primitive, including a blue-
green alga and an anaerobic photosynthetic bacterium. They found the
carbon isotope fractionation in the compounds of the various organisms
to be closely comparable.

Hoering examined suites of rocks, including both modern and ancient
samples in which he could compare carbonate from the matrix with associ-
ated reduced carbon. The difference in C^^/C^^ isotopes between the car-
bonate and reduced carbon was found to be closely comparable throughout
the whole time span. This difference was about the same as the difference
in isotope ratios of growth mediums and that fixed in the photosynthetic
organisms. Studies by others had shown that the major part of the isotope
fractionation occurs at the moment of CO2 fixation in the photosynthetic
organisms.

In summary, the work with biochemical fossils and with carbon isotopes
pointed toward a long enduring continuity in very important biosynthetic
processes.

Astronomy

For most of this century world leadership in observational astronomy
has rested in Pasadena, California. During the first five decades, Carnegie's
Mount Wilson Observatory was the primary center. Later, when the 200-
inch telescope on Palomar became operational, Carnegie scientists partici-
pated with astronomers from the California Institute of Technology in the

REPORT OF THE PRESIDENT 23

joint operation of facilities on both mountains. The combined operation
has continued under the aegis of Hale Observatories. During the past
decade new, advanced equipment of the highest quality has become avail-
able at other sites around the world. Many of these facilities are supported
by governments or consortia that provide several times the amount of
funding available to Hale Observatories. Despite its comparatively small
budget, Hale continues its tradition of excellence at the frontier of astro-
nomical research.

George Ellery Hale, the founder and first Director of the Mount Wilson
Observatory, was relentless in the search for excellence, and his spirit
remains at Pasadena. By a combination of skill, judgment, and intuition
he chose the site at Mount Wilson, where the best seeing conditions in
North America are found. The telescopes erected there were of the best
quality. Hale was also a leader in applying to astronomy the advances in
physics, especially those in spectroscopy.

The equipment and techniques were put to good use by many astrono-
mers, but among them Edwin P. Hubble stands out. We now know that
the universe is populated by galaxies, many of them like our own. But
until recently this fact was not recognized. What are now known as galaxies
were called nebulae. They were bright patches in the sky, usually spiral
in form. But even the nearest one in the northern hemisphere, Andromeda,
was so distant that stars in it were not resolved. Using the superior equip-
ment at Mount Wilson and aided by good seeing conditions, Hubble was
able to identify individual stars in Andromeda. He also found that some
of the stars varied in their intensity, the period being a function of their
luminosity. These '^Cepheid variables" had been studied both in our own
galaxy and in the Magellanic Clouds, and their distance from Earth had
been measured. Thus, it was possible to calculate the distance to the
Andromeda spiral and to note that it was far removed from our galaxy.
From this beginning, the concept of a many-galaxy universe evolved.

The development at Mount Wilson of instrumentation capable of
spectroscopy of very faint sources was helpful to Hubble in making another
discovery — that of the expanding universe. He followed up on an earlier
observation by V. M. Slipher, who had noted a redshift that seemed to be
related to the distance galaxies are from us. After a survey that included
many galaxies, some of them far removed, Hubble concluded that the
universe was indeed expanding. His conclusion raised some basic questions :
What was the universe like earlier? Did it begin with a ''big bang"? How
fast is the universe expanding? Will it continue to expand forever or will
it ultimately reach a maximum size and then begin to contract as gravi-
tation overcomes initial expansion?

Many of these questions are still not settled. One of them — how fast is
the universe expanding? — has been of special interest. The measurement

2J^ CARNEGIE INSTITUTION

of redshifts is not easy. The problem is to establish an accurate distance
scale for far-off galaxies. As one of many accomplishments, Allan R.
Sandage of Carnegie has conducted painstaking work over many years
to improve the distance scale.

Hale's early efforts to apply precise spectroscopic techniques were con-
tinued at Blount Wilson's Pasadena headquarters at Santa Barbara Street.
A machine for ruling gratings was built that supplied the needs of Mount
Wilson and many other observatories around the world. The gratings are
an essential component of the coude spectrograph attached to the 100-inch
Hooker telescope, still one of the best in the world. With this equipment
it was possible to resolve complex spectra of many stars and obtain infor-
mation about the chemical content, temperature, pressure, density, strength
of gravity, electric fields, and the magnetic fields in specific regions of a
star. Other observatories also participated in this effort, but Carnegie was
the leader. The huge body of data accumulated was later employed in the
formulation and testing of models of nuclear synthesis and nuclear re-
actions in the stars.

In connection with his work on nucleogenesis, Paul Merrill at Mount
Wilson made an interesting discovery: the presence of technetium in stars.
Technetium (element 43) occurs on earth only as short-lived radioactive
isotopes made by neutron activation of molybdenum in nuclear reactors.
No stable form exists. Thus the discovery of technetium in stars means
that neutron activation of molybdenum (element 42) is occurring in stars.

Another major long-term study that has made a difference in astronomy
was the series of observations by Walter Baade. He discerned and named
two types of stellar populations in nearby spiral galaxies. The population
II stars appear to have been formed early. They are, for the most part,
distributed in spherical pattern around the galaxy. The population I stars
tend to lie in the plane of the galaxy. Some of them are young stars that
give evidence of having formed in the presence of hydrogen. Baade's
observations gave rise to the suggestion that galaxies are formed by con-
densation of enormous gas clouds. This mechanism is still under active
investigation.

The Hale Observatories had a crucial role in the optical identification
of quasars. Radio astronomers had discovered major sources of radio noise
at various places in the sky. But they could not associate the noise with
any particular object. In 1960 Thomas A. Matthews, a radio astronomer
at Caltech, was able to provide precise locations for some of these radio
sources, and Sandage from the Carnegie side was able to fix upon cor-
responrling optical sources. He saw that they had unusual optical spectra.
Later, Maarten Schmidt of Caltech proved that the reason for the odd
spectrum was an enormous redshift. That is, the objects were at distances
as great as 15 billion light years.

REPORT OF THE PRESIDENT 25

Genetics

The Carnegie Institution has a long history of distinguished involve-
ment in genetics. During the life of the Department of Genetics at Cold
Spring Harbor (1902-1962), Carnegie did more to nurture this science than
any other institution.

The first outstanding contribution at Cold Spring Harbor was the de-
velopment of hybrid corn by George Harrison Shull. Later there was
important activity in genetic mapping and study of mutations and be-
havior of chromosomes in Drosophila. Another notable activity that con-
tinues today is the work in plant genetics conducted for many years by
Barbara McClintock. Her discovery that genes that control other genes
can move around in the chromosomes is today one of the single most
exciting issues of molecular biology. For many years Cold Spring Harbor
was a major center of molecular biology and the site of Nobel Prize-
winning experiments by Alfred D. Hershey and Martha Chase.

More recently, major advances in genetics have also been made at other
Carnegie laboratories. During the 1960s Roy Britten and his colleagues
developed in the Biophysics Section of the Department of Terrestrial
Magnetism new and powerful tools for determining the details of chromo-
some structure. They found that while bacterial genomes consist essentially
of several thousand different genes, eukaryotic DNA consists of families
of more or less related sequences as well as sequences occurring in only a
single copy per genome. At our Department of Embryology splendid
progress has been made in gene isolation during the past decade.

Hybrid corn. George Harrison Shull's development of hybrid corn has
been this century's single most important scientific contribution to agri-
culture. The economic value of the development is now estimated at more
than 100 billion dollars.

In two papers, published in 1908 and 1909, Shull presented recommenda-
tions to the American Breeders' Association. Shull noted that corn farmers
used simple techniques of selective breeding, choosing seed corn each
year from the better specimens of the past crop. This, however, produced
injurious effects in the long run because by repeated self-fertilization a
relatively pure but fragile genetic strain would evolve. Shull suggested
that the corn-breeder, instead of seeking the best pure line, should find
and maintain the best hybrid combination.

First of all, he said, the practical agriculturist should systematically
develop multiple pure-line specimens. This was to be done by repeatedly
self-fertilizing descendents of a single specimen, from generation to genera-
tion, until an essentially pure (homozygous) state was reached for a
number of breed lines. Then, from the multiple lines developed, every pair
combination would be crossed. The qualities of each resulting hybrid

'36 CARNEGIE INSTITUTION

offspring woukl be studied, and the most desirable set of parents identified.
Tlie proiluction of seed-corn could then take place, using a method devised
by Shull. He developed two nearly homozygous specimens, and he showed
tliat cross-breeding them brought a marked improvement m the first-
generation offspring.

At first there was some resistance to Shuirs reconmiendations. One
obstacle was the general view that obtaining the hybrids would be too
costly for widespread use. A few agricultural experimentalists, however,
were impressed by Shull's results, and in 1919 a member of the Connecticut
Experiment Station published details of a double-cross technique involving
foiu' inbred strains. Serious efforts at developing hybrids followed, but
results unavoidably awaited passage of many growing seasons. The first
advertisement for commercial seed-corn appeared in 1926. The use of
hybrid corn spread rapidly during the 1930s, and by 1946 two thirds of
American corn acreage was in hybrid corn. Today virtually all United
States corn comes from hybrid seeds as does much of the corn grown
abroad. Today yields are about triple those of the pre-hybrid era. Part
of the improvement is the result of fertilizers and other factors, but most
of it can be attributed to hybrid seeds.

Shull's work receives credit in every modern writing on the development
and impact of the hybrid corn idea. For this work Shull received the
Marcellus Hartley Medal of the National Academy of Sciences in 1949
"in recognition of his services in the application of principles of the pure
line and of hybrid vigor to the improvement of the quantity and quality
of the maize crop."

The passing years have added to the significance of Shull's discovery.
Success with corn hybrids led to the use of similar approaches to obtain
hybrid vigor in a wide variety of other plants and in some animals.

The professional staff of the Department of Genetics was of modest
size; the usual complement of permanent senior investigators was about
five. However, most of the great names of genetics were in one way or
another connected wdth it. Milislav Demerec, writing in 1942, mentions
some of the geneticists whose work had been fostered by the Institution.
The list of Research Associates included: W. E. Castle, E. B. Wilson,
T. H. Morgan, C. B. Bridges, A. H. Sturtevant, C. E. McClung, R. Pearl,
H. E. Crampton, E. B. Babcock, H. D. Goodale, L. R. Dice, F. B. Sumner,
Th. Dobzhansky, and J. Schultz. Demerec noted ''A particularly strong
measure of support was provided during those early days when genetics
w^as in special need of recognition and assistance. This backing given to
genetical research by the Institution undoubtedly accounts to a large
degree for the fact that the United States now occupies a leading position
in this branch of science."

In the two decades after 1910, the most significant research in genetics

REPORT OF THE PRESIDENT 21

was that of Thomas Hunt Morgan, who studied Drosophila melanogaster
(the fruit fly). He received regular grants from the Institution, and he
published some of his important results in Carnegie monographs and in
its Year Books. Morgan and researchers at Cold Spring Harbor regularly
exchanged comments on one another's work.

In the early 1930s the center of Drosophila work was Cold Spring
Harbor. Demerec carried on studies of x-ray induced mutations and
deficiencies in the animal. Of special interest was the role of gene mutation
in ontogeny. One of the early contributions of the radiation work was the
relation of induced deficiencies to cell lethality.

With the introduction of salivary gland chromosome analysis, the work
on deficiencies and other types of chromosome aberrations advanced rap-
idly, in collaboration first with Margaret Hoover and later with B. P.
Kaufmann, Eileen Sutton, and others. Demerec also examined factors con-
trolling spontaneous mutability. He studied the differences in mutability
in various wild-type strains of D. melanogaster, and identified a mutability-
stimulating gene in the Florida stock.

Demerec and Kaufmann are also known for a small handbook of simple
experiments with fruit flies, published in 1940. Nearly 300,000 of their
Drosophila Guides have been sold to secondary schools and colleges, as
well as to individuals who wanted an introduction to ^^hands-on" genetics
research.

Demerec had unusually fine talents as an administrator. He was quick
to recognize the emergence of new important ideas from whatever source.
He organized the world's best annual symposia on quantitative biology,
which brought together experts from all over the world. Cold Spring Harbor
in the summer was a place of great intellectual ferment.

Salvador E. Luria, one of the founders of molecular biology, has given
his evaluation of the role of Cold Spring Harbor as an intellectual cradle
for molecular biology.

There are starry years in the history of science. For genetics
one such year was 1941. Under new guidance by Milislav Demerec
the Department of Genetics of the Carnegie Institution of Wash-
ington and the Biological Laboratory of Cold Spring Harbor set
in that year one of the foundations on which molecular genetics
was born and successfully grafted upon the stock of Mendelism.
The 1941 Cold Spring Harbor Symposium saw physicists, chem-
ists, and geneticists begin to create and to speak a common
language, just as in the Department of Genetics itself Demerec,
McClintock, Kaufmann, and Fano were breaking down the
disciplinary barriers in the hunt for the gene as molecule.

The trend was not ephemeral. It was rooted in the insight put
forward by Muller and Delbriick that molecular paradigms had
to be substituted for formal paradigms if genetics had to explore

iS8 CARNEGIE INSTITUTION

not only the order of genes but also their nature and their func-
tion. With this in mind Demerec welcomed and cultivated the
buddint!; field of bacteriophage research between 1941 and 1950.
Phage courses and "old" and young phagologists were at home in
Cold Spring Harbor and brought to it bacterial and phage
genetics. Witkin joined the Genetics Department in 1945 and
made it into a center of bacterial radiobiology. (For many of us
sinnmer visitors she was also a patient interpreter of the exciting
but not readily understandable discoveries by McClintock on
generegtilation.)

Bacteria and phage brought to Cold Spring Harbor a crowd
with more biochemical and biophysical savvy: Roberts, Stent,
and many others. Hershey settled in the Department of Genetics
and in 1952 discovered the transfer of DNA from phage to
bacterium and dissipated the last questions as to the nature of
the genetic material. Watson spent in Cold Spring Harbor the
summers of 1948 and 1949 and returned in 1953 with the Holy
Grail whose hunt Cold Spring Harbor had inspired: the molecular
structure of the gene.

The stimulus provided by the intellectual ferment at Cold Spring
Harbor ^vas for the most part manifested in research performed elsewhere.
However, the stimulus was not lost on the small staff resident there. When
Alfred D. Hershey set out to show that the phage nucleic acid (and not
its protein) performed the genetic role, success in surmounting the major
technical difficulties came quickly. With Martha Chase, Hershey tagged
the phage nucleic acid with radioactive phosphorus, its protein with sulfur.
Hershey and Chase first confirmed that the phage injected only its nucleic
acid into its victim, and that the phage protein shell remained attached
to the victim cell membrane. The researchers next employed a centrifuge
and a high-speed mixer in several experiments designed to strip away all
viral protein residue from victim cells. The production of viral progeny
was unaffected. This result demonstrated that DNA is the sole genetic
agent of the phage, and confirmed that nucleic acid was involved in the
heredity of all creatures.

One of the Institution's guiding principles, as set forth by Andrew
Carnegie is: ''To discover the exceptional man . . . and enable him to
make the work for which he seems specially designed his life work."

Few scientists have so magnificently justified the Institution's support
as has Barbara McClintock. She came to be regarded as the world's lead-
ing cytogeneticist, and in 1970 received the nation's top civilian science
award, the National Medal of Science.

In 1967 Marcus M. Rhoades, himself a distinguished geneticist and a
member of the National Academy of Sciences, evaluated her work: ''Among
McClintock's outstanding contributions is her analysis of the control of
gene action in maize and the discovery of the two-unit interacting system.

REPORT OF THE PRESIDENT 29

This concept was the precursor of the regulator-operon theory of gene
regulation that won for its promulgators, Jacob and Monod, the Nobel
Prize in 1965. Her finding that the transposition of controlling elements
from one chromosomal location to another was accompanied by a change
in gene action afforded a new and revolutionary insight into chromosome
structure and genie expression. Genetics would not occupy its present high
estate were it not for McClintock's magnificent and pioneering contri-
butions.''

In the decade after 1967, the significance of this contribution has become
widely appreciated. Others have shown that the type of control system
McClintock noted in maize was universal. The phenomenon has been
popularly called ^^ jumping genes," and among the experts the term trans-
posons is now used.

Much of McClintock's work was done in the late 1940s and 1950s, but
her ideas were so revolutionary that they were slow of acceptance. Her
findings occurred before the discovery of the double helix and the genetic
code. Now it is understood that the phenomena McClintock described
arose from either the insertion of a small piece of DNA into a gene or its
later movement out of the gene.

Rhoades also listed some of McClintock's other accomplishments which
have had major impact on genetics:

Her first major contribution was the demonstration that the
chromosomes were individually recognizable by their relative
lengths and arm ratios, distinctive chromomere patterns, and
deep-staining knobs in characteristic positions. This was followed
by such significant studies as the analysis of translocation
heterozygotes, the correlation of cytological and genetical cross-
ing over, the assignment of linkage groups to specific chromo-
somes, the physical location of gene loci by deficiencies, the
formation of dicentric bridges and acentric fragments as a result
of crossing over in inversion heterozygotes, the somatic and
meiotic behavior of unstable ring chromosomes, the occurrence
of nonhomologous pairing, the structure and function of the
nucleolar organizing region, the production of viable homozygous
deficiencies that stimulated gene mutation and formed a pseudo-
allelic series, and the genetic and cytological consequences of the
bridge-breakage-fusion cycle. Her studies on the evolutionary
history of races of maize as disclosed by the number and location
of specific chromosome knobs have been conducted with typical
precision and elegance.

Department oj Plant Biology

When carbon dioxide is fixed by plants, the process is often written
CO2 + H2O + h/x -^ CH2O + O2. That is, photosynthesis gives rise to

so CARNEGIE INSTITUTION

carbohydrates plus oxygen. Behind the shnple equation is a complex
reality. ]\lany different carbon compounds are produced. Of equal im-
portance are the steps in converting light energy into the chemical energy
that drives the synthetic processes. It is now known that when light falls
on chlorophyll and other pigments, a series of excitations and electron
transfers takes place. Some of the steps occur as quickly as about 10~^^
second; others proceed more slowly. It is also known that for photosyn-
thesis to function at top efficiency two pigment systems must be excited.

Recognition of the latter phenomenon was a crucial step in the under-
standing of photosynthesis. Knowledge of the two-pigment process was
largely derived from work at the Department of Plant Biology. The first
indication of a two-pigment system came from the work of Robert Emer-
son, a Research Associate of the Institution in 1938. He found that the
efficiency of photosynthesis was anomalously low at the far red end of
the spectrum. To explain this finding he suggested that there were two
pigment systems, and that one of them did not absorb at the far red end
of the spectrum. At first, his suggestion was not generally accepted. But
it was later confirmed by C. Stacy French and Jack Myers, who conducted
further experiments with short flashes of light (that is, a short exposure to
far infrared light was followed by a dark period and then by a flash of
light at shorter wave lengths, for example, in the red. Photosynthesis pro-
ceeded efficiently. Thus, the two light reactions could be separated in time.
The observations were extended to a series of photosynthesis organisms
having a variety of pigments, and the phenomenon was shown to be
general.

In most plants the predominant effective pigment is chlorophyll. The
existence of the two-step process with one step effective in the far infrared
suggested that there might be physically different forms of chlorophyll,
each with its own light absorption characteristics. To investigate this
possibility, French designed and built an instrument that permitted him
to measure the first derivative of the absorption spectrum. With this
apparatus he was able to show that there are as many as four distinct
forms of chlorophyll in amounts varying from plant to plant.

A major activity at the Department of Plant Biology has been the study
of the nature and organization of photosynthetic pigments. Harold H.
Strain pioneered in development of chromatographic methods for the
separation of carotenoids and chlorophylls that had been extracted into
organic solvents. Beginning with the work of French, Harold W. Milner,
and others, fractionation of chloroplast materials into active components
has been continued to the present. L. M. N. Duysens discovered a light-
induced absorption change at 515 nanometers in the alga Chlorella which
has led to clarification of the role of carotenoid reactions in photosynthesis.
Victoria H. Lynch and French discovered that carotenoids could be re-

REPORT OF THE PRESIDENT 31

moved from dried chloroplasts by petroleum ether extraction and later
added back in active form. David C. Fork subsequently clarified the role
of the copper protein plastocyanin in the electron transport chain between
system II and system I. Such studies have continued to the present with
the recent work of Jeanette S. Brown on purification and characterization
of reaction center chlorophyll-protein isolated from a wide variety of
photosynthetic organisms.

Another discovery of major importance was the observation by James
H. C. Smith of the light-driven transformation of protochlorophyll to
chlorophyll in extracts of leaves of dark-grown plants. This work, followed
by that of Kazuo Shibata showing absorption shifts in vivo for the newly
formed chlorophyll a, marked the beginning of our understanding of the
biochemistry of pigment changes occurring in chloroplasts of leaves when
they are exposed to light for the first time.

Today laboratories at the forefront of photosynthesis studies use very
sophisticated equipment, including the latest in electronics coupled into
dedicated or general computers. The Department of Plant Biology has
often been a leader in the invention and use of advanced equipment.
French and Fork have contributed to such developments. One of French's
inventions is a comparatively simple but very useful device that forces
cells through a needle valve as a way of breaking them open while leaving
their intracellular structures relatively intact. The instrument is now used
all over the world by plant biologists, bacteriologists, molecular biologists,
and others.

Plants can be grown in a greenhouse and studied in the laboratory, but
by going to nature and studying plants in a wide variety of circumstances,
undreamed of lessons can be learned. Such a program has long been an
important aspect of the Department's research. California is an ideal spot
for this kind of work. Within easy access are the heat of Death Valley,
the frigid temperatures of the Sierras, and the cool, damp environment of
the coastal areas. Taking advantage of some of the opportunities for study,
Jens C. Clausen, David C. Keck, and William M. Hiesey, in collaboration
with Malcolm A. Nobs, performed a remarkable series of experiments in
which they transplanted representatives of a species found in diverse en-
vironments from one environment to another. They discovered that differ-
ences between plants from two different locations were primarily genetic —
not just a function of the cooler or warmer environment affecting growth.
This work and the extensive genetic analysis accompanying it provided
the basis for our knowledge of ecological races as an early step in the
evolution of plant species. The work is discussed in detail in every modern
textbook of plant evolution and finds its way into books on animal evolu-
tion as well.

An outgrowth of the experimental studies by Clausen, Keck, and Hiesey

5J8 CARNEGIE INSTITUTION

was a program to develop improved range grasses using the techniques
oi hybridization and transplantation developed in the studies already
mentioned. With extraordinary thoroughness, these investigators assembled
a massive number of collaborators and had their test species and hybrids
grown all over the United States. The C-1 strain of Poa pratensis discov-
ered at that time now contributes as much as 80% of the seed in commer-
cial lawn mixtures all over the West, Midwest, and Northeast. C-1 stands
for Carnegie number 1.

Another result of the studies by Clausen, Keck, and Hiesey was the
development of a mobile laboratory that now enables Olle Bjorkman in
collaboration with Harold A. Mooney of Stanford University to make
sophisticated studies of photosynthesis and respiration and related gas
exchange on plants growing in severe climates. Field studies are being
conducted in the deep shade of the rain forest and the intense summer
heat of Death Valley. The results are integrated with those of laboratory
studies in controlled-environment chambers, leading to precise characteriza-
tion of biochemical, physiological and physical characteristics that enable
plants to survive under such extreme conditions.

While these studies have been important in showing the significance of
already known adaptations such as C4 photosynthesis, they have also led
to the discovery of new adaptations. For example, James R. Ehleringer
has showm that Encelia species growing under desert conditions use a dense
layer of epidermal hairs to reflect back as much as 70% of the light that
falls on their leaves. By this mechanism, leaf overheating is prevented
witli a minimum of evaporative cooling. Bjorkman and Joseph A. Berry
have found that certain desert plants in situ can tolerate water stresses as
great as — 50 bars without any effect on the quantum efficiency of photo-
synthesis. One of the desert plants, Tidestromia oblongifolia, gives a re-
markable demonstration of thermal stability. The optimal temperature
for its growth is near 46°C (114.8°F), and it can photosynthesize effectively
at still higher temperatures. Under optimum conditions, this plant doubles
its dry matter every two and a half days.

The combination of intensive comparative laboratory and field studies
of plants from extreme environments has been a major program of the
Department for the past ten years. It has provided the foundation for
much of modern physiological plant ecology.

D ppart m (m t of Em bryo logy

An uncommon and important characteristic of Carnegie departments is
the abihty to evolve drastically but in an orderly manner. Programs are
not changed for the sake of change or to follow the fashion of the times,
but because major programs have been completed or because other insti-

REPORT OF THE PRESIDENT 33

tutions appear ready to carry them on. Such circumstances do not arise
very often. When they do, careful thought is given to the directions new
programs will take.

These remarks are particularly applicable to the Department of Embry-
ology. The Department had made superb contributions in the study of
human and primate embryology, but by 1955 the challenges were diminish-
ing and other institutions were entering the field. James D. Ebert and
President Haskins determined that new directions should be followed while
existing work was gradually phased out. Today the Department of Em-
bryology is still at the frontiers of developmental biology, but the program
is completely different from that of two decades ago.

When the Department's work was initiated in 1914, knowledge of human
embryology, particularly that of the first trimester of pregnancy, was in
a primitive state. As a result, the original program of research was centered
on descriptive embryology. Objectives for study included (1) curve of
growth, (2) anatomy of various stages; modeling and dissection of internal
structure, (3) morphology of the brain, (4) histogenesis: the differentiation
of the various tissues, (5) the causes of spontaneous abortion, (6) study
of monsters, (7) study of moles (proliferative malformations of the
placenta and membranes), and (8) comparative and experimental em-
bryology.

By 1955, Ebert 's predecessor, George W. Corner, could say that the
Department possessed a collection of human embryos that was by far
the largest in the world, one that was well-prepared, safely housed, and
kept usable by good records and ample indexes. The statement is still
true, although the Carnegie Collection is now at the University of Cali-
fornia at Davis.

Corner believed that the problem of the curve of growth had been
solved:

We now know the external appearance and dimensions of the
human embryo and fetus day by day from the two-cell stage to
birth. The time schedule of development has been worked out
with a fair degree of certainty, so that the age of an embryo
can be determined from its external appearance with an error of
plus or minus one day, in the first seven weeks, and to the correct
week in later stages. The records and data on this time schedule
include a superb collection of photographs and models, a descrip-
tive catalogue . . . which furnish norms of growth in weight and
length of the body, and various other useful dimensions through-
out gestation.

In the study of the anatomy of growth stages. Corner thought that the
special accomplishment of the Carnegie laboratory had been the accurate

Sj^ CARNEGIE INSTITUTION

aiul dot ailed reconstruction in three-dimensional models of the internal
structure throughout the embryonic period.

Another major accomplishment was the series of studies of the embry-
oloiry oi nonhuman primates. The Department established a monkey
colony wliich made possible comprehensive study of the reproductive
cycle, includinii' menstruation and the cyclic changes in the uterus and
ovaries. Development of embryos in humans and in rhesus monkeys fol-
lows closely parallel courses. The period of gestation is the same, and while
there are small differences in the schedule of comparable events, these
variations can readily be taken into account. Thus, studies of rhesus mon-
keys can be used to fill gaps in our knowledge of human development.

Information derived from monkeys has been particularly valuable in
providing knowledge about the establishment of the placenta and about
blood circulation during labor.

For example, in Year Book 70, Dr. Elizabeth R. Ramsey of the Depart-
ment wrote. '^An amazingly lucky find caught a 10-day monkey ovum at
its moment of attachment to the endometrial surface. The first tropho-
blastic cell is slipping between two maternal epithelial cells. No comparable
stage has yet been found in man. ... At 11 days the trophoblastic plate
is still solid, but a day or so later fluid-filled lacunae begin to appear in it,
some of them containing maternal blood cells, indicating that maternal
capillaries have now been 'tapped.' " The later stages were elucidated in
comparable detail.

Labor is a crucial event in the feto-maternal relationship, and the
quality of placental circulation is of particular importance at that time
because impairment can lead to anoxia and varying degrees of damage.
Ramsey recognized the importance of seeing the blood circulating in living,
intact animals. Ramsey, together with associates at Carnegie and with
]\Iartin Donner and others at the Department of Radiology of The Johns
Hopkins Hospital, used advanced techniques of still and cineradiography
to observe and record the circulation process. They were able to trace the
pathways of inflow of maternal blood and the drainage back into the
maternal systemic circulation. They showed that muscular contractions
curtailed or halted both arterial inflow and return drainage, and they
demonstrated the basic relationship between maternal and fetal circulations.

The new program initiated at the Department by Ebert has already
yielded results of major significance. In 1963, Irwin Konigsberg published
his "Clonal Analysis of Myogenesis," in which he demonstrated that single
muscle cells could divide in culture and differentiate into muscle tissue.
Until that time it was believed that cells in culture lost their specialized
traits, a process referred to as '^dedifferentiation." This erroneous view had
a profound influence on the field because it implied that cell culture
methods were not useful in studying the specialization of individual cells.

REPORT OF THE PRESIDENT 35

Konigsberg showed that when muscle cells were nurtured carefully, they
maintained their specialized traits and could be kept alive for many
generations. Since that demonstration with muscle cells, many kinds of
specialized cells have been maintained successfully in culture, including
liver cells and cartilage cells (also first carried out in the Department),
several kinds of hormone-producing cell lines, nerve cells, and various
kinds of blood cells. Cell cultures are important because they permit study
of cell differentiation and because tissue cultures can be used in practical
applications. Viruses that infect only specialized cells can be cultured for
vaccine production.

Because of its complexity, the cell surface has been a difficult object
for study. Yet it must be analyzed, for it contains some of the most im-
portant secrets of developmental biology. Through its surface the cell
senses its surroundings and transmits signals to the nucleus and cytoplasm.
Cell surfaces selectively take up nutrients, hormones, and drugs in ways
which profoundly influence development and metabolism. To probe cell
surfaces systematically, it is necessary to identify and characterize indi-
vidual components free from the complex array of molecules in the cell
surface membrane. Owing in large part to Douglas Fambrough and his
colleagues, detailed information is available on one such membrane com-
ponent — the acetylcholine (ACh) receptor on muscle membranes. Fam-
brough and associates have analyzed the synthesis and degradation of ACh
receptors and have obtained information about the assembly and metabo-
lism of cell surfaces. From his knowledge of the abundance and distribution
of ACh receptors, Fambrough predicted that they are disturbed in the
disease myasthenia gravis. He collaborated with Dan Drachman of The
Johns Hopkins Hospital in a study that demonstrated this effect, and
Drachman was inspired to carry the matter further to a successful new
treatment of myasthenia gravis.

Traditional genetics involves selecting for mutants of a gene, mapping
them, and analyzing the altered gene function. Animals and plants are so
complex that this approach has been of limited usefulness in determining
how their genes work, especially during embryogenesis. Research in the
Department from 1962 to the present has helped to establish a new kind
of animal genetics, ^^genetics by gene isolation." In 1964, Donald Brown
and John Gurdon found that a mutant of the frog Xenopus was incapable
of synthesizing the RNA of ribosomes. This led to a series of experiments
which first characterized the RNA molecules in ribosomes, then analyzed
their synthesis in embryos, and finally led to the isolation of the genes
for these RNAs. The ribosomal RNA genes of Xenopus were the first
animal genes ever isolated. It was accomplished in 1966 by Wallace and
Birnstiel at Edinburgh. Brown, I. B. Dawid, R. H. Reeder, and their
colleagues have made a series of important discoveries on these purified

S6 CARNEGIEINSTITUTION

genes. The discovery that "spacer" regions are adjacent to functional genes
was first made with ribosomal RNA genes. In 1968, Brown and I. B. Dawid
showed that frog oocytes specifically amplify ribosomal RNA genes. Gene
amplification is now recognized as an important mechanism for gene con-
trol in tlevelopment. In 1971. Brown and his colleagues purified and
characterized a second gene family of known function from the frog. This
gene (gene family) also encodes for a ribosomal RNA called 5S RNA. This
very simple gene has been completely sequenced by Nina Fedoroff and
Brown, and recently the purified gene has been shown to function in a
cell-free system (Brown and Gurdon). Isolated genes permit the study of
gene function and control without traditional genetics. We hope eventually
to understand what switches genes on and off in development and what
accounts for the differential function of genes in specialized cell types.

Department of Archaeology

Mayan archaeology. The Institution made a great contribution to the
study of ancient civilizations by providing substantial support over an
extended period for archaeological explorations. Most of the effort was
concentrated in Middle America.

The Maya w^ere the most brilliant aboriginal people of the Americas.
Their civilization had its beginnings in a primitive farming culture several
centuries before the birth of Christ. During the Classic Period, which
began about A.D. 300, magnificent temple-adorned cities were built, in-
cluding Tikal, Copan, Uxmal, and many other cities of lesser splendor.
After the Classic Period, the stimulus of the foreign Toltec culture helped
produce the great city of Chichen Itza in northern Yucatan, which rose
to prominence about A.D. 1000.

Among the Maya's intellectual triumphs was extraordinary astronomic
and calendrical knowledge surpassing that of any of their contemporaries
on the Eurasian landmass. The Maya produced hieroglyphic inscriptions
and were the only New World people who consistently and accurately
recorded dates. The Maya had no telescopes, of course, only their eyes,
and patience, and careful records. They computed the solar year at 365.2420
days. Present-day calculations, made with the most advanced computers
and chronometers, give a value of 365.2422.

The Carnegie Institution entered the Maya field in 1914 when Sylvanus
G. Morley was appointed a Research Associate. For ten years he explored
the Maya area, visiting practically all the known sites and discovering
many new ones. His explorations greatly enhanced our knowledge of the
territory occupied by the Maya at different periods and the state of devel-
opment of their civilization at those times. Morley was an expert in de-

REPORT OF THE PRESIDENT S7

ciphering Mayan inscriptions, many of which he discovered, and much
of his work was based on these glyphic time-markers.

In 1924, Carnegie's archaeological efforts were intensified. Chichen Itza
was the primary site chosen by Morley at the beginning of the campaign.
The site is accessible and salubrious. It is famous for the number and
architectural distinction of its buildings. When the new venture began,
Morley and the Carnegie Institution established a policy that was uncom-
mon for its time. Instead of building a '^Carnegie Collection" of priceless
artifacts, the Institution's archaeologists handed over the treasures they
found to the host governments.

Morley wanted to make Chichen Itza a lasting and beautiful monument
to the genius of the ancient Maya. This again involved behavior uncom-
mon at the time. To obtain the data he needed would have been simple.
Plazas could be trenched, pyramids torn open, and the ruined temples
stripped of protective mounds. These procedures would have been quick
and cheap. But they would soon have been followed by destruction by
weather and vegetation, leaving a meaningless jumble of stones. Both the
Mexican Government and the Carnegie Institution understood the neces-
sity for care in digging and leaving all cleared structures in a condition to
resist deterioration. Morley's efforts have endured. Recent visitors to
Chichen Itza have remarked on the excellent appearance of the place.
They have also said that the names Sylvanus Morley and Carnegie are
mentioned with reverence.

Major studies of the Maya continued until 1956, although digging
slowed after 1940. The late 1920s and the 1930s were marked by prodigious
efforts at many sites, leading to scores of books, monographs, and articles
containing thousands of illustrations.

In the study of the Maya, the Institution made a difference. It also
served as a model for others in the conduct of archaeological investigation.

Ceramics. Ceramics are some of the most important artifacts of ancient
civilizations. Emil Haury, the distinguished anthropologist-archaeologist,
has said, ^'Ceramics are to the archaeologist what fossils are to the paleon-
tologist." Clays, the raw materials of ceramics, are well-nigh universal,
and the production of simple, useful objects is easy. Consequently, enor-
mous numbers of potsherds have been found. During digs at one Maya
site, 500,000 items were recovered for study. Often, with time, the types
of ceramics produced in a given place changed, and archaeologists use the
physical appearance of objects as an important source of information.

But sometimes, puzzling or misleading specimens are found. Ceramics
makers may have displayed originality or bad technique. Ceramics from
another locality may have been introduced. Indeed, knowledge of such
importations and of their source is useful as an indicator of commerce and
interchange. In principle, much information can be obtained from chemical

S8 CARNEGIE INSTITUTION

and niineralogical analysis of the objects, since the composition of clays
varies from place to place.

Anna Shepard, an archaeologist of the Carnegie staff, recognized the
value of obtaining and using expert knowledge of the physical chemistry
and mineralogy of ceramics. She studied the subject thoroughly, including
in her education a stay at our Geophysical Laboratory, and she spent
considerable time making ceramic objects. Ultimately she distilled her
knowledge into a 414-page book, Ceramics for the Archaeologist. First
published in 1956, it has been reprinted nine times and continues to be
used as the standard text.

FACTORS FOSTERING EXCELLENCE

The ability of the Institution to make a difference did not arise by
accident. That ability is rooted in the precept that has guided the Institu-
tion since its inception — to "encourage, in the broadest and most liberal
manner, investigation, research, and discovery, and the application of
knowledge to the improvement of mankind. . . ."

Acting within this framework, Vannevar Bush, with the concurrence of
the Board of Trustees, set forth the following description of a point of
view that continues to guide our activities :

Our field is fundamental science. And by this I mean that we
seek out the facts of nature and their interrelationships, without
attempting to apply them immediately to the needs and desires
of man for physical comfort, power to control the forces of nature,
or freedom from disease. We seek to acquire and correlate
knowledge, not to participate in its applications.

We know there may he a long interval between our work and
its practical application, that the disconnected relations we find
must often })e joined with those found by many other investi-
gators before a useful pattern emerges. And yet this does not
trouble us; for we are conscious of the problems that the race
will struggle with long after our work has been merged in the
general advance of science, and later generations will finally
complete work we have l)egun.

Most universities would assert that they follow similar policies with
respect to scientific research. But the Institution differs from its academic
counterparts in the way it conducts research. Patterns of our operation
have evolved that make it possible for our people to be considerably more
effective than many others of perhaps equal gifts. One factor is that the
departments of the Institution have tended to be problem oriented in
contrast to the project orientation or discipline orientation customary at
universities.

REPORT OF THE PRESIDENT S9

An important benefit of problem orientation is that it creates a team
spirit. If the problem is complex and requires an interdisciplinary approach,
the necessary colleagues are recruited. From the beginning, the Institution
has successfully fostered work on major problems that were best tackled
by the interdisciplinary approach.

One advantage of the interdisciplinary nature of the Institution's de-
partments is that people can teach one another. When they are working
together on a problem, something from each person rubs off on the other
members of the group. This interaction tends to produce people of consider-
able professional breadth.

Some of the factors that foster creativity are stimulating and appreci-
ative colleagues, the ability to achieve total immersion and the ability
to avoid distracting interruptions.

With our departmental structure we maintain intellectual islands. People
on the islands can, if they wish, interact with others outside, but they
can also withdraw and be free of interruptions for weeks at a time. Each
place has its own particular esprit de corps, its own drive for excellence.
Ability to be creative is enhanced when a group shares a set of enthusi-
asms and compatible values.

The Institution has functioned successfully under a wide variety of
circumstances. It has adjusted to different economic climates, world wars,
and a drastically changing scientific scene.

In the early days, the amount of research in universities and industry
was relatively small. After World War II, an enormous expansion of aca-
demic and industrial research took place. In monetary terms, the impact
of the Institution diminished although its activities in its chosen fields
maintained their excellence.

During the past decade and especially the past five years, the quality
and quantity of the nation's support for fundamental research has de-
teriorated. A marked shift has occurred in industrial research and develop-
ment toward emphasis on quick pay-offs through improvement of existing
products. Government support of research has become increasingly tangled
in red tape with attendant requirements for very extensive documentation
and associated delays. An especially regrettable development has been the
government's virtual abandonment of support of postdoctoral fellowships.
These adverse trends have occurred at a time when it is becoming clear
that in the future trained minds and scientific knowledge will be even
more important to society.

The Institution's programs are geared to meeting educational and
creative challenges. At the same time, much of our activity is in problems
that relate to future human needs.

One example is our Department of Plant Biology's fundamental studies
of plants. There are two paths that the long-term solution to our energy

40 CARNEGIE INSTITUTION

problems can take. One of them is the breeder reactor and the other is the
use o\ the sun. It is easier to collect the sun's energy in plants than it is
to collect the sun's eneriiv via mirrors. What is more, the chemicals that
are made in plants are more useful than the heat energy that the sun
furnishes. There are many ways to make electricity, which fills a small
part of energy needs, but what of the fuel requirements of a mobile society?
What oi the many chemicals that enter all facets of the economy? The
probabilities are that in the long run at least half the world's energy will
come from biomass. Thus, the work at Plant Biology is in tune with long-
term human needs. Fortunately we are in position to move decisively.
Already there are in progress fundamental studies on the photosynthetic
jM'ocess and investigations into mechanisms by which plants adjust to
harsli environments. We have new facihties in an ideal central location
at tlie Stanford campus. If this facility is not already the world center of
plant biology research, it is about to become so. It has all the necessary
characteristics of place and people and leadership.

Another activity that will meet some of the important needs of the
future is work that is going on at the Geophysical Laboratory. For many
years the staff have been studying the rocks of the earth's crust and
mantle. It has emphasized the study of silicates of which the world has
an abundance. In the end, humanity must become increasingly adept in
using abundant materials.

Another aspect of the w^ork going on at the Geophysical Laboratory
holds promise of yielding an understanding of the mechanisms by which
ores are formed. These processes have long been a mystery. The textbooks
of economic geology are full of speculations on how the various concen-
trations of ore have occurred. No single model of ore formation has been
accepted. In part, the problem was that too many puzzles were involved.
The nature of ore-forming fluids was unknown, as well as why they should
move. The means by which particular elements were carried, how they
were concentrated, also was a mystery.

What helps make the study of these matters more feasible at this time
is understanding that has come from the concepts of colliding continental
plate.^. It has become apparent that at least some of the ore-forming
processes are a by-product of the collisions and subduction of plates. Such
collisions and later related magmatic intrusions give rise to necessary
temperatures and permeabilities of rocks that result in motions of fluids.
If we completely understood the ore-forming processes, we would know
how to interpret the outliers that accompany the ores. These are often
found at some distance from the ore-body itself. Once the processes were
understood, a limited amount of drilling could provide a large amount
of information and guidance. The advantage of understanding deep-seated
processes would be substantial, since the cost of drilling deep holes is very

REPORT OF THE PRESIDENT 4^

high. If we are to find the deeper ores, we must have a better understand-
ing of the processes by which they are concentrated.

An important activity at the Geophysical Laboratory is its continuing
development of super-high-pressure equipment in which experimentation
can be carried out under unprecedented extreme conditions. In the past
such pioneering has always led to new insights. Often it has resulted in
unexpected practical applications.

An important characteristic that sets humans apart is a search for
understanding — a need to know. All of our departments, of course, seek
to meet this need, but none is so clearly identified with this effort as the
Hale Observatories. Who can look at the stars on a clear night and not
be touched with wonder?

With the new Irenee du Pont telescope in Chile, and the telescopes
on Palomar and Mt. Wilson, our astronomers have at their disposal some
of the world's best instruments. The effectiveness of this equipment is
being continually upgraded by new and powerful accessories. The use of
electronics in the detection of incident quanta permits subtraction of sky
background and gives an important broadening of the spectrum that can
be detected and measured. Where the optical telescopes were once limited
to a range of about 0.3-0.7 micron, the spread is now about 0.3-100
microns.

The ability to work with the additional wavelengths in the infrared has
created many new opportunities. A National Science Foundation Astron-
omy Advisory Committee, which included participants from the Hale
Observatories, recently listed nearly forty important areas for optical-
infrared investigation. They were grouped under four main headings:
Interstellar Medium, Galaxies, Quasistellar Objects, and Cosmology. Among
the items listed under Interstellar Medium were: formation and miner-
alogy of dust; the dynamical role of magnetic fields; the chemical evolution
of the interstellar medium, in particular the return of stellar material after
processing, and the galactic distribution of H2.

Recommended studies of Galaxies included: determination of mass
distributions in galaxies ; investigation of the problem of the missing mass ;
study of the chemical evolution of galaxies, in particular the development
of composition gradients; observation of events in galactic nuclei; deter-
mination of the source and mechanism of ejected material. Other topics
were relationships to sources of chemically processed material, dust forma-
tion and ejection, infrared excesses in galaxies — their origin and the impli-
cations for energetics, and the evolution of physical properties of galaxies
with redshift.

Quasistellar objects remain a source of puzzles and questions. They
represent one of our best modes of observing events close to the beginning
of time.

J^iS CARNEGIE INSTITUTION

Cosmology deals with such questions as whether the universe is open
or closed — whether it will continue to expand indefinitely. This matter
will be investigated intensively during the coming decade.

Under its new Director, George Wetherill, the Department of Ter-
restrial Magnetism has been reshaping its activities toward a more cohesive
program. Many of the earlier activities will continue, for they have been
fruitful and tliey fit into the desired pattern. Some of the questions being
asked include: How was the solar system formed? How^ was the earth
assembled? Was the earth put together by the assembly of heterogeneous
pieces? How do the rocks beneath the continents differ from those beneath
the oceans? What can be said about the structure of the deep interior of
the earth?

These are questions that have been asked many times, but until recently
little substantive information was available to provide a basis for answers.
But (hiring the past decade the solar system has been investigated using
manned and unmanned spacecraft. It has been possible to compare the
evolution of the earth with that of other planets, and to study and chemi-
cally analyze planetary surface materials older than any found on earth.
Clues about the origin of the solar system have been obtained from
meteorites.

Important advances have also been made in the study of the earth itself.
The plate tectonic revolution has drastically altered our concepts of moun-
tain building; it seems clear that the principal dynamic process determining
the development of the crust and upper mantle is the movement of large
lithospheric plates — the destruction of these plates in oceanic trenches
and the formation of new plates at mid-ocean ridges.

The Department of Terrestrial Magnetism has been active in many
studies relating to these two lines of development and is moving to build
upon them.

To provide a comparative background for understanding our own solar
system, optical and radio astronomical studies are under way bearing on
the origin of stars, particularly observational evidence for supernova-
triggered star formation. Observations on pre-main sequence stars of one
solar mass are planned to provide observational constraints on questions
such as the origin of the sun's slow rotation, the relationship of this to a
possible T-Tauri stage of solar evolution, and the luminosity of the early
sun. These are coupled with theoretical and numerical studies concerning
the solar nebula and early solar system, investigations of possible isotopic
anomalies of lithium using new techniques, and studies of charged particle
tracks and microcraters in meteorites resembling lunar regolith breccias.

In the earth sciences the staff are exploiting the insights afforded by
understanding of plate tectonics to investigate the earth's mantle by means
of a combination of geophysical and geochemical techniques. Discoveries

REPORT OF THE PRESIDENT 4^

SO far lend support to models of the earth's origin whereby the planet was
formed on a long time scale (about 100 million years) by the slow accumu-
lation of large bodies, possibly including some as large as the moon. These
models suggest an initial state of the earth that was thermally, mineralogi-
cally, and probably chemically heterogeneous, and in which the working
out of these primordial disequilibria is likely to represent to this day an
important internal process.

Other research projects currently under way at DTM which will con-
tribute to our newly emerging picture of the earth's mantle include: (1)
Seismological studies of the upper mantle beneath continents, island arcs,
and oceans. (2) Geochemical and isotopic studies of rocks derived from the
mantle. (3) Laboratory studies of diffusion in silicate melts and minerals.

(4) Laboratory studies of attenuation and other seismic parameters in
rocks (joint project with Carnegie Institution's Geophysical Laboratory).

(5) Combined theoretical and experimental studies of the earthquake
source mechanism.

Some of the future of the Department of Embryology has been visualized
by Ebert, and the following is adapted from his predictions:

The techniques of gene isolation will be exploited, proceeding from the
earlier successes in isolating and characterizing the genes. Animal genes
are being produced in large quantities using the new recombinant DNA
technologies. It will be possible to study the way in which sequential
functions are related to spatial arrangements of genes. Such sequences
should give information about the control mechanisms which regulate
gene expression.

Even more direct attacks on gene regulation may be made. It should
be possible to understand the relations of genes to the proteins with which
they interact.

Ultimately it should be possible to faithfully reconstruct animal gene
action in vitro.

The cell membrane presents an attractive set of ^^targets." It serves not
only as a port of entry, facilitating the selective transport of materials
into and out of the cell but also as a site of organized biochemical activity.
The central role of membranes in cellular organization has focused atten-
tion in recent years on the structure of the membrane and the relation of
this structure to function and properties.

While it is apparent that the protein components of cellular membranes
play the important structural and functional roles, it has become increas-
ingly clear that the influence of the membrane lipid may be considerable,
both in maintaining a suitable local environment for particular membrane
proteins and in contributing to the spatial organization of membrane
elements through lateral phase separation of membrane lipids.

Just as it has been necessary to isolate and characterize specific genes,

J^ CARNEGIE INSTITUTION

SO it will be essential to isolate specific functional membrane components,
for example, the specific "receptor sites" at which other cells or molecules
interact with a given cell type. The field is a continuum in which we must
take account of the biogenesis and turnover of membrane macromolecules ;
membrane genetics and the differentiation of functional components in
specialized cell membranes; the molecular basis of membrane changes in
response to hormones; and the basis of changes related to cell differenti-
ation and growth, both normal and abnormal.

Within the next decade we should have a far better understanding than
we have today of the structure, functions, and biogenesis of both genes
and cell membranes. We should know the intimate workings of cells that
we perceive today only in bold outline. We should have begun to lay the
groundwork for a concerted attack on the mechanisms whereby cells inter-
act, leading to an attack on the factors that shape tissues and organs
like the brain.

EDUCATION

This report has emphasized practical achievements and research ac-
complishments. Little space has been devoted to educational activities.
Nevertheless, w^e are as proud of our fellowship program and our alumni
as we are of the more visible achievements.

The Institution has fostered the intellectual development of young
people and aw^arded fellowships for study at the departments. Soon after
Caryl Haskins became president in 1955, he expanded the fellowship
program, and the fruits of this initiative are now being seen in distin-
guished alumni. The program was very successful during the late 1950s
and 1960s even while federal funds were readily available for fellowships.
Developments of the last decade have made the program even more
desirable. Government support of fellowships has been drastically cut.
Most postdoctoral stipends that are available are made part of grants or
contracts and bear the title of research associate. What should be a fellow-
ship with the challenges and freedoms that the word implies is instead a
job of work with a definite employer-employee relationship.

At the Carnegie Institution our policies and circumstances encourage
maximum development of the fellows. Equipment, materials, and counsel
are readily available. The fellows are not just part of a team; they are
expected to apply initiative, imagination, judgment, and energy to the
problems they have chosen.

President Haskins' decision in 1955 had an important bearing on events
in 1970. Legislation had been enacted that placed restrictive burdens on
foundations. But, because the Institution had established itself as an edu-
cational institution, it was granted the tax status enjoyed by universities.

The Year in Review

DEPARTMENT OF EMBRYOLOGY

The regulation of cell surface properties during development and the
role these properties play in the interactions between cells are among the
most important phenomena of biology. To study these phenomena one
might choose among many systems. Douglas Fambrough and his colleagues
have made an excellent choice: the relationship of nerves and skeletal
muscles. That relationship is of obvious importance to most forms of life.
The system is a particularly apt one for study because such diverse crea-
tures as leeches, flies, mice, and men use many of the same chemicals and
enzymes in their neuromuscular interactions. Since disruptions of the nor-
mal interactions between nerves and muscles are associated with many
debilitating diseases in man, this parallelism among organisms is of great
potential value as a route to the discovery of cures for human afflictions.

A great array of experimental techniques has been brought to studies of
the nerve-muscle system in the past few years. All the latest developments
in embryological methods have been applied: radioactive tracers, heavy
isotopes, ultracentrifugation, enzymes, and laser beams. This research has
focused to a large extent on the acetylcholine receptors of skeletal muscle
fibers, including the role of acetylcholinesterase.

The acetylcholine receptors are glycoproteins embedded in the plasma
membranes of skeletal muscle fibers. They function as the recognition
system for signals from motor nerves. The number and distribution of
acetylcholine receptors in skeletal muscle fibers are regulated during de-
velopment and in the continuing interactions between nerve and muscle.
An important aspect of this regulation is control of the biosynthesis and
degradation of receptor molecules in the muscle fibers. It has been one of

45

Jfi CARNEGIE INSTITUTION

Fambrough's major research efforts to elucidate the mechanisms of re-
ceptor metabolism.

In this year's Report Fambrough and P. N. Devreotes summarize studies
which indicate that the lipid intermediate pathway of the protein gly-
cosylation (involving dolichol phosphate) operates early in receptor bio-
synthesis. Newly synthesized receptor molecules, already containing
significant carbohydrate, are located in the Golgi apparatus where they
may reside for about 2 hours before transport to the cell surface. They
constitute about 10% of all receptors in the system. In the plasma mem-
brane the receptors are targets for the cells' degradation mechanism, which
involves interiorization of receptor molecules, transport to secondary lyso-
somes, and proteolytic destruction. John M. Gardner has investigated
receptor degradation and found that the normal half-life for a receptor in
the plasma membrane of tissue-cultured embryonic skeletal muscle is about
17 hours.

Another glycoprotein prominently involved in neuromuscular inter-
actions is the enzyme acetylcholinesterase. This enzyme is manufactured
by the muscle in several different molecular forms, occurring variously as
a cytoplasmic enzyme, a cell surface enzyme, and an enzyme released from
the muscle fibers into the extracellular milieu. Richard Rotundo has
directed attention to this enzyme as a likely means of getting at the
mechanisms of nerve-muscle interaction during the formation of neuro-
muscular connections. The study of the acetylcholinesterases has been
conducted using methods developed during the study of acetylcholine
receptor metabolism. These methods facilitated the determination of the
interrelationship of the' molecular forms of the esterase. Comparisons of
receptor metabolism and esterase metabolism may illuminate the relation
between the biosynthesis of plasma membrane glycoproteins and the bio-
synthesis and secretion of secretory glycoproteins.

When considering interactions of cells, a relevant matter is the nature
of their respective membranes. Are the components of a membrane fixed
spatially or are they free to move about? Considerable evidence has been
adduced that some components are mobile.

Fambrough and Richard E. Pagano have begun a collaborative investi-
gation of the freedom with which various surface constituents of cells
move in the plasma membrane. They attach fluorescence molecules to the
cell surface and bleach a tiny region of it with a fine laser beam. The
rate at which the bleached region is then filled is determined by the
fluidity of that portion of the membrane. They have found that ACh
receptors, unlike other regions of the membrane, are anchored to the
surface.

Most of the studies of nerve-muscle interactions at the Department have
been directed toward the muscle component. Kenneth J. MuUer has been

REPORT OF THE PRESIDENT 4^

devoting his attention to the nerves. For his experimental subject he has
chosen the medicinal leech, which has a relatively simple nervous system
in which changes can be followed in detail. When a neural connection is
severed, regeneration usually follows, with the proper connection being
reestablished. Muller describes the sprouting and directional growth of
axons from a crushed neuron toward the distal stump of a neighboring
neuron. The target stump sends detectable electrical transmissions toward
the regenerating fiber until contact is made; then the stump disappears.

One of the goals of the Department has been to isolate genes and study
their activities in vitro. This objective has often seemed unlikely of attain-
ment, but recently substantial progress toward it has been made. The first
breakthrough was the isolation of small amounts of the gene 5S DNA
from the frog. The amounts obtained were very small, but once small
amounts of a gene are available the new recombinant DNA techniques
can be used to replicate the genie material many times. This is accom-
plished by inserting the gene into a specific chromosome of the genome of
the K-12 strain of Escherichia coli and then culturing the altered bacteria
so as to produce a large number of them, and thus of the desired gene. By
the use of established restriction enzyme techniques, the animal gene
material can be extracted from the bacterial chromosome and the genes
then isolated.

This technique was employed by Donald Brown and colleagues to pro-
duce substantial quantities of frog 5S DNA. Subsequently, in a collabora-
tion involving Brown and John Gurdon of the MRC Unit of Molecular
Biology at Cambridge, England, the 5S DNA was injected into the nuclei
of frog oocytes. It was observed that the added 5S DNA had been tran-
scribed by the apparatus within the nucleus to form 5S RNA. This result
seems important, for one can visualize two further lines of experimentation.
The oocyte system could be used as a testing ground either for studying
the action of genes or for determining what components are essential for
DNA transcription.

In other work related to the 5S DNA gene, Nina Fedoroff has used
techniques pioneered by F. Sanger of the MRC Unit in Cambridge to
work out the nucleotide sequence of one full repeating unit of 5S DNA
from Xenopus laevis. This is the first animal gene and its spacer regions
to be sequenced entirely. The sequence says a great deal about the evolu-
tion of the spacer region and helps to localize possible control regions in
the DNA.

In examining the behavior of genes in making specialized substances,
Yoshiaki Suzuki and Brown have devoted considerable effort — and with
good success — to studying the production of fibroin by the silkworm.
In 1972 they isolated the messenger RNA for the protein silk fibroin.
Suzuki, Patrick Gage, and Brown then used the mRNA as a probe to

48 CARNEGIE INSTITUTION

show that cells that do not express the gene have the same number of
genes as those that do express the gene. They found that in the case of
silk fibroin the gene number is one per set of chromosomes. Since that
time, tliis superlative developmental system has been explored by a series
of workers in the Department. These studies have culminated recently
in the isolation of the silk gene by Suzuki with the recombinant DNA
methodology. There was little hope of isolating the gene from the animal's
own n\A, since it is present in only one copy per haploid set of genes
(one part in 40,000 of the animal's DNA). Only indirect studies could
be carried out on this gene. Now, milligram amoimts of the gene are
available in pure form. DNA regions on either side of the gene are present
in some of the cloned gene fragments. These should contain sequences
that control the expression of the silk gene in the living cell.

DEPARTAIEXT OF PLANT BIOLOGY

During the past two years a remarkable increase in the tempo of edu-
cational activities and research has taken place at the Department of
Plant Biology. Located in the middle of the Stanford University campus
and having new laboratory space and new equipment, the Department
interacts with Stanford botanists and students and with scientists of other
universities. The research activity now consists of four major programs:
photosynthesis, membranes, plant DNA, and physiological ecology. These
programs interact and sustain each other, so that the sum is more than
the total of the constituent parts. For example, photosynthesis involves
chloroplasts and membranes. As another example, the ability of some
plants to withstand and thrive in high temperatures is related to the
thermal stability of their photosynthetic apparatus. The inability of some
plants to thrive in temperatures close to but above freezing is related to
the congealing effects of cold on the membranes.

In general, techniques and understanding cultivated by one group are
often utilized by the others. The DNA group, which is quite new, is devel-
oping a body of information that may one day give it an important part
in this interaction. One of the long-range goals is to improve the char-
acteristics of plants. To do so will necessarily involve some kind of change
in their DNA composition. First, however, one must look at the genomes
of existing plants. Later the detailed information that is gathered about
photosynthesis, membranes, and physiological ecology will provide criteria
for choosing which DNA changes to make.

Support of these important activities, particularly those related to
photosynthesis, has been significantly bolstered by a much-appreciated
grant from the Andrew W. Mellon Foundation.

REPORT OF THE PRESIDENT 4-9

The physiological ecology group has done extensive experimentation
on the effects of temperature on whole plants or plant parts. Ulrich
Schreiber and Joseph A. Berry, working with leaves of plants with large
differences in heat tolerance, have monitored the changes in fluorescence
either with gradually increasing temperature or with a sudden temperature
jump under conditions yielding either fixed or variable fluorescence. They
were able to rank the plants by heat stability and to distinguish differences
in heat tolerance in specimens of a single species grown at different tem-
peratures. They also showed that high light intensity can provide sub-
stantial protection against heat damage.

H. A. Mooney, Olle Bjorkman, and James Collatz studied photosynthesis
in Larrea divaricata seedlings from Death Valley grown under three differ-
ent temperature regimes in growth chambers. They found that the tem-
perature optimum for photosynthesis was higher in plants that had been
grown at higher temperatures, and established that the difference in be-
havior was unrelated to differences either in the C02-fixing enzyme or in
some aspect of photorespiration. They examined the effect of water stress
on the quantum yield of photosynthesis for Larrea plants grown under
ample water supply. When the stress was — 36 bars, the quantum efficiency
was less than half that of amply watered plants. In contrast, the same
plants growing in Death Valley showed a normal quantum efficiency and
little change between — 20 and almost —50 bars. The results show that
Larrea can photosynthesize effectively under remarkable drought stress in
nature, but they also prove that one cannot extrapolate from greenhouse
and growth chamber experiments without careful field studies.

Paul A. Armond, Schreiber, and Bjorkman isolated chloroplasts from
the Larrea plants grown under the three sets of temperature regimes, and
investigated fluorescence properties of the chloroplasts with heating. Plants
grown at the highest temperature regime showed little change in fluores-
cence spectra with heating, although they had the largest photosynthetic
units. Chloroplasts from plants grown on the other two regimes, however,
showed marked heat damage to the mechanism of energy transfer from
chlorophyll b to chlorophyll a at temperatures well below those required
to inactivate electron support.

More detailed experiments by Schreiber and Armond showed heat dam-
age to the transfer of energy among forms of chlorophyll a as well; but
during heat inactivation those reaction centers which were still active
were fully active, and the photosynthetic unit size unchanged. These
studies have added substantially to our knowledge of how one plant copes
with drought stress and of the mechanisms of heat damage w^hen it occurs.

In another study on heat damage, Bjorkman and Murray R. Badger
investigated the thermal stability of 14 different photosynthetic enzymes
from the high-temperature-adapted C4 species Tidestromia ohlongijolia

50 CARNEGIE INSTITUTION

ami the cool-temperature C4 species Atriplex sabulosa. Of the 14 enzymes
studied. 4 had identical or similar heat stabihties in the two species, while
another 6 showed differences of 6-10°C.

Badiier. Aaron Kaplan, and Berry have also continued work begun last
year on the mechanism by which the alga Chlarnydomonas reinhardtii
adapts to growth on low concentrations of CO2. Their results show that
the adaptation involves development of a system to concentrate CO2 from
the external medium. This system is unrelated to C4 photosynthesis in
higher plants.

Fork has continued his studies of the influence on photosynthetic activi-
ties of the physical phase transitions of thylakoid membrane lipids. In
chloroplasts of lettuce and spinach that have high concentrations of un-
saturated fatty acids, these phase transitions occurred well below 0°C.
Fork and Xorio Murata obtained evidence that increasing either the
magnesium or the potassium concentration in chloroplasts from spinach or
lettuce markedly depressed phase transition temperatures. In contrast to
the lettuce and spinach, in which the phase transition temperatures were
all below 0°C, the thermophilic blue-green alga Sy7i echo coccus lividus —
capable of growing photosynthetically at 75°C — showed a phase transition
near 43°C when grown at temperatures above 55°C. When the algae were
grown at lower temperatures, lowered phase transitions were seen. Con-
sistent with the results obtained earlier with the blue-green alga Anacystis,
the photosynthetic electron transport capacity of Synechococcus declined
sharply below the phase transition temperature.

]\Iurata and Fork also found that phase transitions could be detected by
measuring the dark decay of the light-induced carotenoid change. These
studies showed that chilling-sensitive plants had a phase transition above
0°C that varied with growth temperature, while chilling-resistant plants
did not.

Fork and Mordhay Avron, who is on sabbatical leave from the Weiz-
mann Institute in Tel Aviv, studied the influence of temperature on the
rate of proton efflux through thylakoid membranes as monitored by the
fluorescence of 9-aminoacridine. They found the rate sharply temperature
dependent and an excellent indicator of phase transitions and other
temperature-dependent membrane parameters. This technique also demon-
strated differences between chilling-sensitive and chilling-insensitive higher
plant chloroplasts.

Avron and Schreiber describe use of the 9-aminoacridine for monitoring
proton pumping and efflux, to determine how closely proton pumping
parallels changes in the reduction of the primary system II electron
acceptor when reverse electron transport is driven by exogenous ATP.
Under a wide variety of conditions, the buildup of protons in the thylakoids
was remarkably parallel to the reduction of the primary acceptor, strength-

REPORT OF THE PRESIDENT 61

ening the chemiosmotic hypothesis for energy transduction in chloroplasts.
In addition, they found conditions under which electron transport could
be driven backward to the photosystem II reaction centers with ATP, and
they were able to monitor the process by chlorophyll luminescence.

Larry N. Vanderhoef, on sabbatical leave from the University of Illinois,
collaborated with Briggs in completing a detailed study on the influence
of red light on elongation of the first internode of germinating corn
seedlings. Red light treatment brings about an inhibition of growth that
can be reversed by adding the plant hormone auxin. The inhibition shows
two phases, one in response to extremely low doses of light and the other
requiring far larger doses. An action spectrum for the first phase shows
that there is a peak at about 665 nm, indicative of the action of the well-
known photomorphogenic pigment phytochrome, but there is a second
sharp peak at 640 nm which cannot be attributed to phytochrome. Earlier
workers had noted the peak but otherwise ignored it. In membrane frac-
tions from corn, Steven J. Britz, Suzanne Widell, and Briggs have detected
a pigment absorbing near 630 nm that is effectively bleached by both red
and blue light. They present evidence that it is a reasonable candidate
for the 640-nm photoreceptor (pigments frequently show spectral shifts
upon extraction). These studies point toward the existence of a photo-
morphogenically active photoreceptor other than phytochrome absorbing
in the red region of the spectrum.

In other work on pigments, Robert Brain, Dow Woodward, and Briggs
obtained evidence that Neurospora mutants deficient in 6-type cyto-
chromes were also deficient in light-reducible cytochrome in membrane
fractions and showed significantly impaired photosensitivity. That the
reduced photosensitivity was not simply an indirect result of lowered
respiratory capacity in the mutant was suggested by the appearance of
normal photoreducible cytochrome responses in membrane preparations
from another kind of respiratory mutant with normal cytochromes. These
results, together with some obtained with pea and methylene blue, are
the strongest to date implicating the flavoprotein-cytochrome complex in
the blue light photoreception process.

HALE OBSERVATORIES

Anyone acquainted with investigators at the Hale Observatories must
be impressed with the scope of their activities. A small staff, including ten
members on the Carnegie side and eleven Caltech members, operate the
telescopes on three mountains. They conduct deeply scholarly studies that
are at the forefront of astronomy. They develop many new forms of elec-
tronic and other instrumentation. They .act as hosts each year to some 60

5f CARNEGIE INSTITUTION

guest investigators from many institutions in this country and abroad.
During the past seven years, staff have also devoted considerable attention
to the design, construction, and testing of the new Irenee du Pont tele-
scope on Las Campanas in northern Chile. Completion of this instrument
removes a burden as it creates new opportunities for the staff.

The design of a modern telescope that meets high standards of excel-
lence necessarily entails many thousands of decisions, with a corresponding
number of ways in which something can go wrong. Manufacture of the
components also involves many occasions for errors. In addition, the
logistical effort required to bring everything together on a remote moun-
tain like Las Campanas is enormous. The task of building a telescope on
Las Campanas is less demanding than that of building one on the moon,
but not very much less. Everything that is needed must in both cases be
brought to the site over a long supply line.

Thus the staff, especially Horace W. Babcock, Director, and Arthur H.
Vaughan, Assistant Director for Las Campanas, deserve warm congratula-
tions on bringing the telescope into operation.

Excellent photographic plates have been obtained from the instrument,
and it is now being used for astronomical research. Additional auxiliary
equipment is still being installed, but the major tasks are completed.

The quality of the first large photographic plates is particularly im-
pressive. The instrument and the accompanying plate-holder were designed
for square plates 51 cm on a side, with star images that are no larger than
0.25". The high quality of the plates attests to the overall excellence of
design and construction of the telescope. The quality of the plates also
ensures that the instruments will be very valuable in sky surveys.

Alembers of the Hale staff enjoy certain traditional and organizational
advantages. A small organization can be especially flexible and responsive
to new opportunities. Hale astronomers enjoy many nights a year of tele-
scope time, so they are able to undertake long-range research projects and
address fundamental questions of the cosmos which can only be answered
by the accumulation of large amounts of data.

One such program is the one now being brought to completion by Allan
Sanflage and G. A. Tammann on the observed velocities (redshifts) of all
the brighter galaxies. It has involved the compilation of a consistent body
of observational data over a forty-year period. The effort was begun in
1936 by Hubble in his contacts with N. U. Mayall of the Lick Observatory
and in direct collaboration with Milton Humason at Mount Wilson. Its
aim was the acquisition of data either by direct spectroscopic observation
or by compilation from published sources, to give the redshifts of all of
the 1249 bright galaxies of the vShapley-Ames catalogue. That catalogue,
published by the Harvard College Observatory in 1932, lists positions,
angular sizes, and estimated magnitudes. A 1956 paper by Humason,

REPORT OF THE PRESIDENT ' 53

Mayall, and Sandage summarized results of the program up to that time.
New spectroscopic observations, including many obtained by Sandage in
Australia in 1969 and others he made at Palomar between 1970 and 1977,
have resulted in 670 new redshift values for bright galaxies. Sandage and
Tammann, in collaboration, have now compiled all known redshifts for
the galaxies in the catalogue. It is expected that the revision will be pub-
lished by the Carnegie Institution in 1978.

The revised Shapley-Ames catalogue will include the reclassification
as to type (spiral, elliptical, irregular, etc.) of all listed galaxies using the
Hubble system. This reclassification was initiated by Hubble in 1946 on
the basis of improved direct photographs. Over the years, Sandage has
accumulated plates obtained with the 5-meter Hale telescope in the north
and with the 1-meter Swope telescope at Las Campanas in the south.
These photographs, together with others in the collection of plates made
with the Mount Wilson telescope, form the basis for the classification of
the more than 1200 new galaxies appearing in the upcoming publication.

The immediate purpose of Sandage and Tammann in compiling the
redshift observations is to provide a data sample that will permit deter-
mination of the sun's motion (and of the motion of our Galaxy) relative
to the inner metagalaxy — '^our part of the Universe." Analysis of the new
body of data, with its nearly complete velocity coverage, has been begun
by A. Yahil of Tel Aviv University and Tammann, with the aim of
measuring perturbations of the local velocity field in the presence of
density contrasts. Their goal is to assess the role of gravity in determining
the global world model by using the velocity perturbations as measures
of the local ratio of kinetic energy to gravitational potential energy.

The subject of the structure and evolution of galaxies is now very active,
and significant new insights are being achieved. A few years ago, Leonard
T. Searle reported the discovery of chemical composition gradients across
the disks of spiral galaxies. The discovery has been confirmed, and atten-
tion has turned to the origin of the gradients and to the detailed physics
of the ionized regions whose spectra provide evidence for these gradients.
In the past year, Searle has collaborated with G. A. Shields of the Uni-
versity of Texas in analyzing new observations of emission regions in the
spiral MlOl. Searle and Shields find that in addition to the well-established
abundance gradient there is also a gradient in temperature of the ionizing
stars, the hottest stars occurring far from the center of the galaxy. Assum-
ing that the ionizing stars share the abundance gradient of the gas, they
show that this has important effects on the character of the ionizing
radiation emitted by these stars. When these effects are both taken into
account, the detailed models produce an excellent fit to the new observa-
tions, and Searle and Shields find that the 0/H abundance ratio falls off
as r~^^^ when r, the distance from the center of the galaxy, lies between

54 CARNEGIE INSTITUTION

5 ami 25 kiloparsecs. The observations rule out a theoretical prediction
that the abundance would decrease linearly with radius.

Stephen A. Schectman and George W. Preston have initiated a search
for stars oi very low metal abunchmce in the halo of our Galaxy. The
discovery of such objects would place important constraints on the early
history oi nucleosynthesis during the contraction phase. For this survey,
they have equipped the 4G-cm Schmidt telescope at Palomar with a large
interference filter and an objective prism. The filter limits the extent of
each spectrum to 100 A around the H and K lines of ionized calcium. This
technique, with a survey limit near the 14th magnitude, permits easy
rejection of the vast majority of stars showing normal H and K line
strengths. The residual group of rarer objects includes w^hite dwarfs, nuclei
of planetary nebulae, and the very-low-metal-abundance stars that are
the objects of the survey.

The use of the SIT Vidicon Cassegrain spectrograph and the multi-
channel spectrometer has resulted in the discovery of over 50 additional
degenerate stars by Jesse L. Greenstein and collaborators. They continue
to find evidence for stratification in the atmospheres of degenerate stars
and for the presence of relatively sharp but weak hydrogen lines in the
spectra of quite cool objects that are dominated by helium. Two red
degenerate stars studied by Greenstein and W. F. van Altena of Yale
I'niversity are among the intrinsically faintest known, with visual lumi-
nosity of about lO"** that of the sun.

The new application of heterodyne spectroscopy at submillimeter wave-
lengths has been started at the 5-meter telescope by T. Phillips and T.
Huggins of the Bell Telephone Laboratories and Gerry Neugebauer and
Michael W. Werner of the Hale Observatories. The receiver, based on an
InSb hot-electron bolometer mixer, is mounted at the prime focus of the
telescope. Its use has resulted in the first detection in an astronomical
source of the ./ = 3-^/ = 2 line of CO at 345 GHz. The line has been
mapped over the central few arc-minutes of the Orion molecular cloud.
The emission peak is within 10" of known infrared sources.

Jerome Kristian, Sandage, and James A. Westphal are continuing their
program for measurement of the redshifts and magnitudes of the brightest
cluster galaxies to extend the Hubble diagram to large redshifts. They
now present new photometry for 33 clusters and new redshifts for 50
clusters, extending to ^ = 0.75. The new data, combined with earlier sam-
ples, show evidence for a positive curvature of the Hubble diagram, with a
formal value for the deceleration parameter go (uncorrected for evolution)
of 1.6.

To avoid possible selection effects for the largest redshift clusters,
Kristian, Sandage, and Westphal restricted the analysis of their sample to
clusters with z < 0.4. The currently available data samples show, for the

REPORT OF THE PRESIDENT 65

first time, a significant departure of the Hubble diagram from a straight
hue, with persistent indication of a positive curvature. If no evolutionary
corrections are applied, the formal best-fit value for qo is 1.6 ± 0.35. The
dispersion in the absolute magnitudes of the brightest cluster galaxies
continues to be strikingly small: The best-fit values are Mv = —23.28 and
Mr — — 24.09, with a dispersion of 0.28 mag for the first-ranked cluster
galaxy in each passband.

After a lapse of nearly fifty years since Bubble's original discovery of
the redshift-distance relation, and the extension of observed distances by
a factor of 200 since then, the data appear to be near to fulfilling their
original promise as a possible test for cosmological models. The validity
of this approach for finding go is now dependent on the outcome of efforts
to understand the nature and size of galaxy-evolution effects during the
look-back time. To consider two opposite cases, the present data show that
if the universe is nearly empty {q^ ^ 0), then galaxies must have been
about 1 mag brighter at z =^ 0.4. On the other hand, if there has been no
net brightness evolution since z = 0.4, then the data show that the universe
is firmly closed, finite, and oscillating.

Althea Wilkinson and J. Beverley Oke have completed a study of the
absolute spectral energy distributions of 54 brightest galaxies in faint
clusters. About 15 objects have redshifts between z = 0.10 and 0.21 and
constitute the sample with short look-back time. The remaining 39 clusters
have redshifts between 0.21 and 0.47. Over the range in time corresponding
to ^ = 0.10 to 0.46, B — V is found not to change by more than 0.03 mag.
Other effects of size comparable to or greater than any color evolution
are present, and a comparative analysis of the individual spectral-energy
distributions suggests that one significant effect may be a dispersion in
metallicity by a factor of 4 among the galaxies in the sample.

DEPARTMENT OF TERRESTRIAL MAGNETISM

To the huge populations that live in areas such as the Pacific rim,
earth tremors are a continuing menace against which most inhabitants
feel resigned and helpless. To Selwyn Sacks at DTM, earthquakes pre-
sent great challenges. He joins with others in seeking ways of predicting
the occurrence of quakes; but he also employs seismic waves as a tool to
study the structure of the earth, including the deep interior, the roots of
continents, and the properties of subducting tectonic plates.

In such studies he uses information from other scientists from seismic
arrays such as the WWSN (Worldwide Seismic Network). He supplements
these data with observations made with very effective instruments invented

56 CARNEGIE INSTITUTION

and built at DTM, and he maintains his own inexpensive but productive
international seismic array.

Japan, the People's Republic of China, the Soviet Union, and the
United States are carrying on large programs aimed at earthquake pre-
diction. To date, the most useful program has been that of the Chinese.
However, they completely missed predicting the huge quake of July 28,
1976. Evidently the relatively simple models of events preceding earth-
quakes leave out important phenomena. One of these is ''slow earthquakes,''
a phenomenon described in this Report by Sacks, Shigeji Suyehiro, Alan
T. Linde, and J. Arthur Snoke. These are extremely low frequency seismic
events characterized by large strain energy but radiating almost no energy
in the normal frequency band. They are not detected by ordinary seis-
mometers but can be observed with the borehole strainmeter developed by
Sacks and Dale W. Evertson. This instrument extended the spectrum of
measurable seismic frequencies to zero Hertz.

Using evidence obtained from borehole strainmeters located in Japan,
Sacks and his associates now suggest that slow earthquakes may play a
major role in the redistribution of strain energy preceding a major earth-
quake, and that this redistribution may not occur until shortly before
the earthquake. Because of this, it may not be possible to predict large
earthquakes until a few days before they occur, since the necessary high
stresses may not be present in the earthquake source region before then.

In another study Kiyoshi Suyehiro and Sacks report measurement of the
velocity structure of a slab of lithosphere in the Japan arc that is undergoing
subduction into the interior of the earth. Observations like these are
making it clear that the subducted slab is more than a hypothetical entity
required to satisfy conservation of mass. It is a real and mappable feature
of the earth. The new measurements supplement earlier investigations
of the slab by Sacks, Hiromu Okada. and others that showed the slab to
have a higher seismic velocity and Q than the surrounding asthenosphere,
and identified the boundary between the slab and the underlying astheno-
sphere. The new work shows that the seismic velocity within the subducted
slab is not constant, but increases toward the lower boundary of the slab
and can be interpreted in terms of a two-layer model of the slab. This
model is in general agreement with the thickness and structure of the
normal oceanic lithosphere in the western Pacific, and shows that for the
most part these characteristics are preserved as the lithosphere undergoes
subduction.

Recent work is showing that the distinction between continent and
ocean is a fundamental one and that a continent is characterized not only
by its superficial crust but by a characteristic upper mantle as well. The
continental lithosphere is found to be hundreds of kilometers thick, in
contrast to the much thinner oceanic lithosphere. This is most apparent

REPORT OF THE PRESIDENT 67

in South America, where Sacks and Snoke identified seismic velocity re-
versal at a depth of 400 km as marking the hthosphere-asthenosphere
boundary.

In their present report, Sacks, Snoke, and Husebye extend this con-
clusion to the Baltic shield. Using the Norsar seismic array in Norway,
they collected data on the seismic arrival designated Sj) (waves from
distant earthquakes which propagated through the mantle as shear waves
and converted to compressional waves at the continental lithosphere-
asthenosphere boundary). From the arrival times of this phase, they
calculated a boundary depth of 250 ± 15 km. Furthermore, the polarity
of this converted Sp arrival relative to the normal shear wave arrival was
used to show that the velocity decreased with depth at this boundary.
This strengthens the interpretation of the boundary as one between a rigid
lithosphere and a partially melted underlying asthenosphere.

When combined with surface wave data, Q structure, and heat flow
studies, this picture of a distinctive continental mantle of great thickness
seems likely to be valid. Moreover, last year Christopher Brooks, David
E. James, and Stanley R. Hart interpreted isotopic data from this and
other laboratories as indicating that the continental lithosphere was old as
well as thick. In this regard it seems similar to the deeper oceanic mantle
beneath the asthenospheric source of the abundant ocean ridge basalts,
which also shows isotopic evidence of preserving heterogeneities over a
major portion of the earth's history.

Our solar system was formed about 4.6 billion years ago, some 10 billion
years or more after some of the earliest stars were born. The chemical
composition of the planets provides evidence that the material of the
solar system was once part of an antecedent star or stars. Studies of
meteorites have shown that they too were assembled from materials that
a relatively short time before had been part of a stellar nuclear reactor.
What happened in the few millions of years before the solar system was
assembled?

One hypothesis that has long been considered is that new stars are
formed from the remains of old ones. This year substantial evidence has
been accumulated in support of this suggestion. William Herbst and George
Assousa report both radio and optical evidence that the formation of stars
is triggered by the shock waves associated with the explosion of super-
novae — stars in the final stages of their evolution. They describe a region
in Canis Major that they have identified as a supernova remnant by its
circular structure and by 21-cm radio observations of an expanding hydro-
gen cloud of a kind characteristic of supernova remnants.

Another likely candidate of this type is also described by Assousa, Herbst,
and Kenneth C. Turner. The discovery of even a few such objects is a
strong argument that the process is an important one in star formation.

58 CARNEGIE INSTITUTION

Most identifiable supernova remnants are so young that star formation
cannot yet have occurred in them, while in most known regions of star
formation the process is so advanced that any evidence of an earlier
supernova in the region has disappeared.

George Wetherill has carried out new numerical and theoretical work
on the important role played by relatively small planetary bodies in solar
system events. The most complete quantitative treatment of the accumu-
lation of a cloud of planetesimals into a terrestrial planet like the earth
has*been given by Safronov and his co-workers. This earlier work considers
the case in which planets are assumed to move in circular orbits, with
negligible mixing of material from the vicinity of one planet to that of
another. The present work extends this to the case of multiple planet
growth with a planet's eccentricity, inclination, and distance from the sun
allowed to vary as it will during its growth, in conformity with the laws
of conservation of energy and angular momentum. It is found that con-
siderable mixing of material should occur between one terrestrial planet
and another : if the terrestrial planets did actually accumulate in this way,
it is unlikely that their composition can be interpreted simply in terms
of condensation from a solar nebula at temperatures determined by their
solar distances, as is now commonly believed by cosmochemists.

In a separate report, an investigation is made into the fate of the ^'left-
overs" from this planetesimal swarm after the planets have grown to nearly
their final mass. It is found that this evolutionary history may provide
a natural explanation of the characteristics of the ''late heavy bombard-
ment" of the moon and terrestrial planets which ended about 3.9 billion
years ago, long after the formation of the planets themselves. Some
members of this residual swarm may still be observable in the innermost
asteroid belt. A third study, coauthored with J. G. Williams, describes
why these particular asteroids are likely to be the present-day sources of
iron meteorites as well as of meteorites which have experienced igneous
differentiation, particularly the basaltic achondrites.

GEOPHYSICAL LABORATORY

In chemical laboratories, under carefully controlled conditions, remark-
ably quantitative separations of constituents can be obtained. Elements
originally present in very low abundances can be isolated in essentially
pure form. On the earth's surface, in the oceans, and in the interior of the
earth, enormous quantities of material are chemically processed by nature.
On the surface of the earth, rocks and soils are selectively leached by
water containing carbon dioxide and organic acids. In the oceans, sub-
stances arriving from land are exposed to a higher pH and salinity than

REPORT OF THE PRESIDENT 59

that normally found on land, and reactions and sedimentation occur. In
parts of the ocean, substances are exposed to anoxic environments con-
taining hydrogen sulfide and organic chelating materials. Under these
conditions, a great deal of chemical processing and concentrating of ele-
ments occurs.

In the interior of the earth, partial melting of rock and formation of
magmas occur with partial chemical separation. When the magmas solidify,
particularly when they cool slowly, there are further concentrations.

Important mechanisms that give rise to ore deposits of some elements
occur relatively close to the surface of the earth. These processes involve
the movement of very hot water containing dissolved substances through
rocks that have been partially fractured. Some constituents of the rocks
are more readily dissolved than others. Near the surface many events
occur; for example, as the solutions cool, solid materials are deposited.

Large volumes of the earth's crust have participated in one or more
episodes of weathering and subsequent sedimentation in the oceans. Some
of the sediments have been subducted as part of tectonic plates. Thus, the
history of the chemicals of the crust has been very complex. The scientific
approach to assembling a basis for judgment on such complex phenomena
is to break the larger problem into more manageable pieces.

Qualitative features of the foregoing processes are well known ; what has
been missing are many of the quantitative details, especially those involved
in the formation and cooling of magmas and in the processes occurring
when hot fluids interact with crustal materials ultimately to form ore
deposits.

These phenomena have come under increasing attention at the Geo-
physical Laboratory. This year's reports describe many studies of the
partitioning of trace elements during partial melting and solidification.
They also summarize new studies aimed at working out the nature of
ore-bearing fluids.

This year Ho-Kwang Mao and Peter M. Bell used the diamond anvil
pressure cell to produce pressures as high as 1,500,000 bars in experiments
simulating conditions in the upper mantle of the earth. Their experiments
revealed that perovskite is the most common structure found at these high
pressures.

In the diamond cell, Ho-Kwang Mao and Peter M. Bell determined the
specific volumes of Cu, Mo, Pd, and Ag to approximately 1 Mbar. The
specific volumes were referred to the pressure-volume Hugoniot data
obtained in shock waves. These metals were selected because their Hugo-
niot data require little correction and because they are considered the best
available material in terms of physical properties for studies of equations
of state. In addition, the spectral shift of the ruby Ri fluorescence line
was measured in each experimental run ; thus the volume data on the four

60 CARNEGIE INSTITUTION

metals constitute primary pressure calibrations of the ruby spectral shift.
The ruby fluorescence shift can now be used as a secondary pressure
calibration.

Mao. Takehiko Yagi. and Bell have begun an extended investigation of
the experimental mineralogy of the deep mantle with the diamond cell.
They report quenching experiments in the system Mg-Fe-Ca-Al-Si-0 in
which new products could be identified by x-ray diffraction. The experi-
ments done at 100-450 kbar and 1000°-1200°C produced complex assem-
blages dominated by phases with the perovskite structure. One long-
standing prediction about the mineralogy of the deep earth (300-1200 km)
was that the known minerals in the upper mantle would be converted
mainly to perovskite structures in the lower mantle. The results of the
new experiments confirm that perovskite is a major phase. The possibility
exists that all mineral compositions invert in part to perovskite structures
at high enough pressures, but the present results indicate that the assem-
blages will be more complex. The study includes in situ x-ray diffraction
measurements to identify nonquenchable phases.

Yagi. Mao. and Bell systematically studied the end members of the
common rock-forming minerals in rocks at high pressures to ascertain the
compositional limits of the perovskite structure. They were successful in
synthesizing a perovskite structure, essentially a single phase, from the
enstatite composition, MgSiO.^, and in determining its structure. The struc-
tural determination led to a number of conclusions that may prove useful
in understanding the behavior of other minerals at the high pressures of
the earth's mantle.

First, the metal-oxygen bond distances are significantly greater in the
more dense perovskite than in pyroxene, olivine, or the oxides — perovskite
has closer packed oxygen atoms. The metal site is very large, and the
crystal-field intensity is greatly reduced. This reduction particularly affects
the partitioning of iron, causing it to be excluded from the perovskite
structure.

Because of this, the perovskite structure is probably not as dominant in
the mantle as would be predicted from density considerations alone. This
leads to the suggestion that in the deep mantle perovskite and mineral
oxide phases are coexistent.

In another area of high-pressure experimentation, Robert M. Hazen
anrl Larry W. Finger have modified a diamond-anvil cell for use on a four-
circle, single-crystal x-ray diffractometer. The modified cell permits the
determination of crystal structure while the crystal is at pressures of up
to 100 kbar, making it possible to relate the physical properties of a
crystal to its structure at the same pressure. The modified high-pressure
cell combined with a new crystal-mounting technique and data collection
procedure has resulted in the collection of a wider array of diffraction

REPORT OF THE PRESIDENT 61

data, better resolution, and a higher pressure range of greater stabihty.
Studies on pyroxene, ruby, sihcate spinels, phlogopite, and a chlorite were
carried out with the modified cell.

In high-pressure refinements of the fassaitic clinop5n:'oxene from Angra
dos Reis meteorite, Hazen and Finger have demonstrated the anisotropies
in bond compression that exist in some silicates. In general, larger poly-
hedra with cations of lower valence compress more than smaller polyhedra
with cations of higher valence. Furthermore, longer bonds in large poly-
hedra compress more than shorter bonds. On the basis of these observa-
tions, they suggest that the packing of oxygens in clinopyroxene becomes
more regular at high pressure.

Finger and Hazen refined the crystal structure of a synthetic ruby
doped with 0.4 mole % Cr at pressures up to 80 kbar. This pressure is
considerably higher than had been previously obtained with a single-
crystal, diamond-anvil high-pressure cell. The compression is linear within
experimental uncertainty; however, the compressibility of the c axis is
50% greater than that of the a axis. Under compression, the longer Al-0
distance decreases more than the shorter one does; thus the structure
becomes more regular as the pressure is increased.

In most of the studies made on rock-forming materials the substances
are subjected to high temperature and high pressure for a day or more.
The materials are then very quickly cooled (quenched). Often the chemical
compounds existing under the extreme conditions remain essentially un-
changed during the quick cooling. However, this is not always true and
major questions remain about the properties of silicates at high tempera-
tures and pressures. Two studies are particularly noteworthy. One showed
that the chemical nature of a molten silicate containing iron was sub-
stantially altered by pressure. The other showed that when a fraction of
a rock melts there is a simultaneous and very large change in its electrical
conductivity and in the transmission of sound waves through it.

David Virgo and Bjorn 0. Mysen employed Mossbauer techniques to
study changes in the valence state of iron, particularly the effects of
pressure on the redox states of iron in the melt. They noted a disappear-
ance of part of the Fe^+, which they attribute to a change in the degree
of polymerization of the silicate melt.

The presence of liquid in a rock may have a major effect on its physical
properties. For example, the decrease in seismic velocities in some regions
of the upper mantle has been attributed to the presence of a small amount
of melt. To examine this effect, Tsutomu Murase, Ikuo Kushiro, and
Toshitsugu Fujii measured the compressional wave velocity in a peridotite
in the region of its beginning-of-melting temperature at 1 atm. The
velocities observed were almost constant at temperatures below that of
the solidus but dropped rapidly to values close to those of molten basalt

62 CARNEGIE INSTITUTION

at temperatures just above the solidus. The variation of volume of melt
with temperature and the aspect ratio of the thin films of melt observed
confirmed that just a small amount of partial melt had had a considerable
efi'ect on the wave velocity.

Simultaneous with the change in w^ave velocity at the point where
peridot ite bciiiins to melt, a large increase in electrical conductivity was
noted at both 1 atm and 15 kbar. This sharp increase may be useful for
marking the solidus temperature.

LOSSES

Although he will not retire until the end of fiscal 1977, I want to
acknowledge in this Report the contributions of Horace W. Babcock to
astronomy and to Carnegie Institution.

He was born and grew up in Pasadena, California, not far from Mt.
Wilson Observatory, w^iere his father worked as a physicist. He took his
B.S. degree from the California Institute of Technology and his Ph.D.
from the University of California. After serving as an instructor at the
Yerkes Observatory in Wisconsin and the McDonald in Texas, he moved
to Boston to do research on radar at M.I.T.'s Radiation Laboratory. The
next year. 1942, he returned to Caltech for a project in rocketry.

After the War, Dr. Babcock joined the Mt. Wilson and Palomar
Observatories (now^ Hale Observatories) as a Staff Member of Carnegie
Institution. His first project there was a search for magnetic fields in sharp-
line stars. To detect the fields, he designed and built the Zeeman analyzer.
This sensitive instrument enabled Babcock to find strong magnetic fields in
a great many stars. He was then able to classify them according to the
variability of their field strength and the reversal of magnetic polarity.

Later, Dr. Babcock's focus shifted to the magnetic fields over the surface
of the Sun. With his father, he devised the solar magnetograph, an electro-
optical system for measuring and recording the strength, polarity, and
distribution of the fields. Dr. Babcock also found ways to improve the
ruling machine that prepared diffraction gratings, and he supervised a
ruling laboratory where the ruled gratings were for many years produced
and distributed as a service to astronomers around the world.

When the 200-inch telescope was being erected on Palomar, he designed
a precise automatic guider for steering it during stellar observations.

Since 1964 Dr. Babcock has directed the Observatories during a period
of great challenge and opportunity. Perhaps the two outstanding achieve-
ments of the Observatories during Babcock's administration were the
cosmological studies done by Sandage, Schmidt, and others, which changed
our conception of the Universe; and the establishment of a major optical
installation in the southern hemisphere, at Las Campanas, Chile.

REPORT OF THE PRESIDENT 63

The Institution's first Distinguished Professor, Elburt F. Osborn, re-
tired this year. Dr. Osborn's career in petrology has included two periods
of association with our Geophysical Laboratory. He was a Staff Member
at the Laboratory from 1938 to 1945. And in 1973 he returned as Dis-
tinguished Professor of Carnegie Institution. In the intervening years
Dr. Osborn was Professor of Geochemistry at the Pennsylvania State
University, where he also served as Dean of the College of Mineral Sciences
and Vice President for Research. In 1970 he was appointed Director of
the U.S. Bureau of Mines of the Department of the Interior, a post he
held until rejoining Carnegie Institution in 1973.

Dr. Osborn's work on hydrothermal studies and crystallization phe-
nomena and on phase equilibria has generated more than a hundred publi-
cations and has brought him numerous honors, including the Roebling
Medal of the Mineralogical Society of America, the John Jeppson Award
of the American Ceramic Society, and four honorary degrees.

. . . AND GAINS

Maarten Schmidt, 47, has been chosen by Carnegie Institution and the
California Institute of Technology to direct the Hale Observatories, be-
ginning July 1, 1978.

Dr. Schmidt is best known for his recognition of the large redshifts in
the spectra of quasars, a finding that led to the conclusion that they are
the most distant and most powerfully radiating objects in the Universe.
Quasars continue to be an important part of Dr. Schmidt's research pro-
gram, along with radio sources of other kinds and x-ray sources. Recently
he has been studying stars that are relatively close to us — only about a
hundred light-years away.

Dr. Schmidt's association with the Observatories began in 1956 when
he came to Pasadena as a Carnegie Fellow, after taking his Ph.D. in
astronomy at the University of Leiden. After his fellowship he went home
to The Netherlands but returned to the U.S. in 1959 to become a Staff
Member of the Observatories and an Associate Professor at Caltech. He
has served as Executive Officer for Astronomy and Chairman of Caltech's
Division of Physics, Mathematics, and Astronomy. In his new post of
Director of Hale Observatories, Dr. Schmidt will continue as Professor of
Astronomy.

In Maarten Schmidt the Hale Observatories will have an outstanding
scientist of international reputation and an experienced administrator to
carry on a tradition of excellence in astronomy.

The following honors were awarded to Staff Members during the past
year:

^4 CARNEGIE INSTITUTION

George W. Wetherill, Director of the Department of Terrestrial Mag-
netism, was elected President of the International Association of Geo-
chemistry and Cosmochemistry at the Association's General Assembly
in Paris in May 1977.

L. Thomas Aldrich of DTM was reelected General Secretary of the
American Geophysical Union for a two-year term that began in July 1976.

A'era C. Rubin of DT]\I was appointed a member of the Visiting Com-
mittee of the National Radio Astronomy Observatory and a member of
the Msiting Committee of the Department of Astronomy, Harvard Uni-
versity.

Winslow R. Briggs. Director of the Department of Plant Biology,
was elected President of the California Botanical Society for the year
1977-1978. He was also elected a Fellow of the American Association for
the Advancement of Science.

James E. Gunn and George W. Preston of the Hale Observatories were
elected to the National Academy of Sciences.

Peter M. Bell of the Geophysical Laboratory received the Exceptional
Scientific Achievement Medal at the 1976 NASA Annual Honor Awards
Ceremony in November 1976.

During the past year the Institution has enjoyed the generosity of its
trustees, employees, and friends, who contributed a total of $241,302 to aid
the Institution in its work.

Grant support has been equally gratifying. The Andrew W. Mellon
Foundation has endowed the Institution with $200,000 a year for the next
three years to be used for research in photosynthesis at the Department
of Plant Biology. The Max C. Fleischmann Foundation has given a
quarter-million-dollar grant for renovating and equipping the building in
which the Embryology Department is housed in Baltimore. The Carnegie
Corporation of New York has agreed to extend fellowships in the natural
sciences for the next four years, and is providing $90,000 a year for this
purpose. All the departments of the Institution will benefit from this
grant. In the past year the National Science Foundation provided addi-
tional grants of $340,000 which went to DTM, Hale, and the Geophysical
Laboratory. NASA gave grants in the amount of $114,000 for research at
Hale Observatories, and the Embryology Department benefited from
grants totaling $353,000 from the U.S. Public Health Service. Other grants
were received from the Whitehall Foundation, the Helen Hay Whitney
Foundation, the Office of Naval Research, the Muscular Dystrophy Asso-
ciation, and the Corning Gla.ss Works Foundation.

REPORT OF THE PRESIDENT

65

FACULTY, FELLOWS, AND STUDENTS

1976-1977

DEPARTMENT OF EMBRYOLOGY

Baltimore, Maryland

Director

Donald D. Brown

Staff Members

Igor B. Dawid
Douglas M. Fambrough
Kenneth J. Muller
Richard E. Pagano
Ronald H. Reeder
Yoshiaki Suzuki

Fellows

Peter Botchan
Salvatore T. Carbonetto
Diana Card
Jeffrey Doering
Mark Dworkin
Scott Emmons
Nina Fedoroff
Paul Geshelin
Elizabeth Godwin
Lawrence J. Korn
Eric Long
Yasumi Ohshima
Keiko Ozato
Richard Rotundo
Alex Sandra
Barbara Sollner-Webb
Masatoshi Takeichi
Katherine Tepperman
Yoshihide Tsujimoto
Walter Wahli
Harvey Wahn
Peter K. Wellauer

Students

Jeffrey Chernak
Brian Cooley
Peter Devreotes
John M. Gardner
Margie M. Goldberg
Robert A. Hipskind
Les Katzel
Marc C. Krauss

Jose Ramirez
Shigeru Sakonju
Nancy A. Union

GEOPHYSICAL LABORATORY
Washington, D.C.

Director

Hatten S. Yoder, Jr.

Carnegie Institution
Distinguished Professor

Elburt F. Osborn

Emeritus Research Associate
Emanuel G. Zies

Staff Members

Peter M. Bell
Francis R. Boyd, Jr.
Felix Chayes
Gordon L. Davis
David H. Eggler
Larry W. Finger
John D. Frantz
P. Edgar Hare
Thomas C. Hoering
T. Neil Irvine
Ikuo Kushiro
Ho-Kwang Mao
Tsutomu Murase
Douglas Rumble III
David Virgo

Fellows

Nicholas T. Arndt
Timothy M. Benjamin
John M. Ferry
Anthony A. Finnerty
Toshitsugu Fujii
Robert M. Hazen
John I. Hedges
Bjorn O. Mysen
Howard R. Naslund
Robert K. Popp

66

CARNEGIE INSTITUTION

Joseph L. Ritchey

Frank S. S}H\ir
Juergon Trochimczyk
Xoreen Tiiross
E. Bruce Watson
Richard F. Wendlandt
Takehiko Yap

Students

Michael J. DeNiro
.TiiUa A. Dill
Dora Y. Lee

Catherine A. McCammon
Marie H. Slezak

HALE OBSERVATORIES
Pasadena, California

Director

Horace AV. Babcock

Fellows

William M. Adams
A. Oer de Briiyn
Alan M. Dressier
Daniel Y. Gezari
Richard F. Green
Ediiardo Hardy
]\Iark Hartoog
Gordon J. Hurford
Vincent Icke
Frank Israel
Stephen L. Knapp
Daniel Kunth
Kenneth A. Marsh
Ronald Moore
Larry D. Petro
Douglas 0. Richstone
Jack W. Sulentic
Trinh X. Thuan
Althea Wilkinson
Peter Wilkinson
Robert J. Zinn

Associate Director
J. Beverley Oke

Staff Members

Halton C. Arp
Eric E. Becklin
Jesse L. Greenstein
James E. Gunn
Robert F. Howard
Jerome Kristian
Robert B. Leighton
Guido Miinch
Gerry Xeugebauer
S. Eric Persson
George W. Preston
Allan R. Sandage
Wallace L. W. Sargent
Maarten Schmidt
Leonard T. Searle
Stei)hen A. Shectman
Arthur H. Vaughan, Jr.
James A. Westphal
Harold Zirin

Staff Associates

Robert J. Brucato
Michael W. Werner

Carnegie-Chilean Fellows

Guido Garay
Maria Teresa Ruiz

Student Observers

Steven V. W. Beckwith
Kirk Borne
Todd Boroson
France Cordova
David J. Diner
Jonathan H. Elias
John G. Hoessel
John P. Huchra
Steven Kent
Barry J. Labonte
Jorge Melnick
Daniel Nadeau
William C. Priedhorsky
Douglas M. Rabin
Russell 0. Redman
Anneila I. Sargent
Donald P. Schneider
William L. Sebok
David Sholle
Richard J. Terrile
Richard Wade
Peter J. Young

REPORT OF THE PRESIDENT

67

DEPARTMENT OF PLANT
BIOLOGY

Stanford, California

Director

Winslow R. Briggs

Staff Members
Joseph A. Berry
Olle Bjorkman
Jeanette S. Brown
David C. Fork
Malcolm A. Nobs
William F. Thompson

Emeritus

C. Stacy French
William M. Hiesey

Fellows

Paul A. Armond
Mordhay Avron
Murray R. Badger
Heather S. Belford
Michael R. Blatt
John E. Boynton
Steven J. Britz
G. James Collatz
John W. Cross
James R. Ehleringer
Arthur W. Galston
George B. Johnson
Aaron Kaplan
Ulrich C. Knopf
Jacob Levitt
John M. Mackenzie, Jr.
Norio Murata
Michael G. Murray
Peter H. Quail
Charles E. Rogler
Ulrich Schreiber
Erich L. Schrott
Alan J. Stemler
Larry N. Vanderhoef

Students

Robert D. Brain
Richard Cuellar
Mary Enama
Holly Gorton
Richard Preisler
U. M. Suzanne Widell

DEPARTMENT OF
TERRESTRIAL MAGNETISM
Washington, D.C.

Director

George W. Wetherill

Distinguished Service Member
Merle A. Tuve

Emeritus

Scott E. Forbush
Richard B. Roberts

Staff Members

L. Thomas Aldrich
George E. Assousa
Louis Brown
Dean B. Cowie
W. Kent Ford, Jr.
Albrecht W. Hoffmann
David E. James
Alan T. Linde
Vera C. Rubin
I. Selwyn Sacks
Fouad Tera
Norbert Thonnard
Kenneth C. Turner

Research Associates

Mordeckai Magaritz
J. Arthur Snoke
Kiyoshi Suyehiro

Fellows

Charles L. Bennett (trainee)
Gregory S. DeWitt (trainee)
John R. Evans
William Herbst
Felix J. Lockman
George H. Pepper
Charles J. Peterson
R. Sundar Rajan
Douglas 0. ReVelle
Stuart J. Weidenschilling
William M. White
David J. Whitford

Reports of Departments
and Special Studies

Department of Embryology

Hale Observatories

Department of Plant Biology

Geophysical Laboratory

Developmental Biology Research Group

Department of Terrestrial Magnetism

Department of Embryology

Baltimore, Maryland

Donald D. Brown
Director

•S ^ ^ ^ ^

^^ c» Q '^

'g ?^ 15 o"
S ,5 .^ :t^

T3

o

bfl

O

m

CD

^ce o rt "2

+= "^ Ph o a

- 3 W o C^

03 W g g ^

03
03

S "

- CO

O "2

CC

O

o

o

bC O o3

"m ^ pq

S a ^ §• .-
^ cc - ^
o ,
;h o3

e^

o

+^
o

cc

S 'a; d
o ^ ^

1— r " "—I r3

'a gwlm I

W)^ > ^ .S^ -5

o3

S <3 ^ CQ

=^ .or.

i.gflil

" i2 e «> c a

Contents

Introduction 7

Studies on Skeletal Muscle and Neuronal Plasma Membranes 9

ACh receptor biosynthesis 11

Studies on the glycosylation of ACh receptors 12

Newly synthesized receptors reside in the golgi 14

Regulation of acetylcholine receptor metabolism by electrical stimulation 17

Degradation 18

Incorporation 19

Degradation of acetylcholine receptors 22

Positional stability of acetylchohne receptors at the neuromuscular junction 28

Studies on chick muscle acetylcholinesterase 29

Properties of chick muscle acetylcholinesterase 30

Cell surface versus internal AChE 32

Release of AChE by chick muscle cells in culture 34

Density shifting of acetylchohnesterase 36

Acetylcholine receptors and a-bungarotoxin binding sites 37

Organization and Interactions of Membrane Lipids in Mammalian Cells 41

Studies of transbilayer distribution of phospholipids in mammalian cells 42

Plasma membrane derivatives 42

Asymmetry of the major phosphohpid species (PC and PE) 42

Acyl chain asymmetry 45

Adhesion of phosphohpid membranes to Chinese hamster fibroblasts:

role of cell-surface proteins *..... 46

Temperature dependence of vesicle uptake 50

Influence of T(, on exogenous lipid incorporation by cells in suspension 50

Mode of vesicle-to-cell adhesion 52

Binding and capping of fluorescent dye containing lipid vesicles to

lymphocytes 53

Regeneration in the Nervous System : Formation of Specific Synapses in the Leech ... 55

Regeneration of an electrical synapse 56

The electrical synapse between S-cells 56

Regeneration of the S-cell axon 56

Reformation of the svnapse 57

"Fusion" \ 58

Survival of the distal stump 61

The distribution of synapses of identified sensory neurons 63

Structural and Functional Studies of the Fibroin Gene 63

Cloning of fibroin gene plasmids 63

Characterization of fibroin gene plasmids 67

Definite identification of fibroin gene in the cloned plasmids 67

Screening and restriction mapping of the fibroin gene plasmids carrying both

mRNA coding sequence and flanking sequence (s) 68

Determination of orientation of the cloned fibroin genes 71

Identification of the site and sequence of the primary transcript of the

fibroin gene 73

Sequencing of a presumed regulatory region of the fibroin gene 75

Ribosomal DNA and Related Sequences in Drosophila melanogaster 75

Studies of cloned rDXA repeats 75
Ribosomal DXA insertions in the Y chromosome nucleolus organizer of

Drosophila 77

Sequences homologous to ribosomal insertions occur elsewhere in the

Drosophila genome 78

Biogenesis of Mitochondria 82

Mapping of 4S RNA genes in A''. laevis mtDNA 82

Ribosomal Genes and their Proteins 84

Structure of the nontranscribed spacers between ribosomal genes 85

In vivo sites of RXA chain initiation on rDNA 87

DXA replication origins in Xeiiopus rDNA 91

Ribosomal gene chromatin 91

Immunological cross reaction between calf and Drosophila histones 94

Genetics bv Gene Isolation: The Dual 5»S RNA Gene System in Xenopus 96

Xlt 5S DNA ' 97

The oocyte injection system 97

Sequencing studies on the 5aS DNA of Xenopus laevis 100

The Structure of X. borealis oocyte and somatic 5S DNAs 102

X'ucleotide sequence anah'sis of the region preceding the Xenopus borealis

5S rRXA gene . ^ 105

Computer analysis of nucleic acid sequences 106

Preparation and transcription of bromodeoxyuridine-substituted Xenopus

laevis 5S DNA 107

The Collection of Human Embryos 108

Developmental stages in human embryos 108

Development of the nervous system 108

Staff Activities ." 109

Bibliography 110

Personnel 112

Carnegie Institution of Washington Year Book 76, 1976-1977

INTRODUCTION

A director is expected to ''chart a tion? Some p;enes make highly specialized

course" for his department. A department substances, while others function in all

with the title of Embryology would seem cells. What distinguishes these two classes

to be already charted in its course. Yet of genes?

the research in this department has Developmental biology is not a field or

evolved remarkably. At present most of a discipline in the same sense as math,

us use molecular tools, but this does not physics, or even some biological areas

acurately convey our direction. Our di- like genetics. Developmental biology is a

rection is wherever talented people take set of questions. The questions deal with

us. Our course is charted by the staff and biological change and the influences on

our colleagues, who follow their own sci- this change. It can safely be said that

entific intuition. My major responsibility these questions are being attacked by

as Director is to support these people in methods of such power that we can expect

every way that will enhance the quality answers in the kind of molecular detail

of their research. not even imagined ten years ago. The

This is not an original approach. I reader of this report will find that many

learned it from Jim Ebert. For 20 years of these new methods are in use in the

the Department has benefited from the Department. Several of the methods have

leadership of a man with broad scientific been developed, at least in part, here,

interests and remarkable organizational In this report Suzuki describes the iso-

ability. Jim Ebert emphasized the quality lation of the gene for silk fibroin with

of research and favored independence and the aid of the new powerful recombinant

innovation. I recognize with gratitude the DNA methodology. There was little hope

role that he has played in encouraging of isolating this gene from the animal's

my own research, in an atmosphere de- own DNA because it is present in only

void of bureaucracy and irrelevant dis- one copy per haploid set of genes (one

tractions. part in 40,000 of the animal's DNA).

Despite the variety of projects, ani- Only indirect studies could be carried out
mals, and materials, there is underlying on this gene. Now milligram amounts of
continuity among the questions being the gene are available in pure form. DNA
asked in the Department of Embryology, regions on either side of the gene are
How are processes regulated in develop- present in some of the cloned gene frag-
ing organisms? What influences a nerve ments. These should contain sequences
to grow along a path toward another which control the expression of the silk
nerve and then form a synapse at just gene in the living cell,
the right moment? How does muscle A second, equally powerful method is
know that it is innervated and respond DNA sequencing. Techniques pioneered
by regulating the abundance and distri- by F. Sanger of the Medical Research
bution of receptors on its surface? What Council (MRC), Laboratory of ]\Iolecu-
surface molecules are responsible for lar Biology in Cambridge, England, have
cell-cell recognition, and how is external made possible the step by step analysis
information that impinges on cells trans- of the arrangement of the nucleotide
lated into physiological change? Why building blocks in DNA. Nina Fedoroff
does a gene work in one cell type but not describes the sequence of one full repeat-
in another? Is the gene itself modified, ing unit of 5S DNA from Xenopus laevis.
or is it directed by the molecules that This is the first animal gene and its
associate with it? What are these mole- spacer regions that have been sequenced
cules, and what regulates their produc- entirely. The sequence says a great deal

8 CARNEGIE INSTITUTION

about the evolution of the spacer region ward the degenerating fiber until contact

and helps to localize possible control re- is made; then the stump disappears,

gions in the DNA. Sandra and Pagano have used an in-

Before the control of any gene can be genious method to measure the content

reconstructed, the true "initiation" site of various lipids on each face of a cell's

of RXA transcription must be known; plasma membrane. AVhen a cell engulfs

and there must be a way to reconstruct a latex bead, the membrane surrounding

faithful gene transcription in vitro. Prog- the bead is inside out. The composition

ress toward both goals is reported here, of the inside surface has been compared

Reeder has made use of an enzyme com- with that of the outside, which can be

plex that can place a "cap" on the first analyzed in an intact cell. Translating

nucleotide in an RNA molecule. He gives asymmetric lipid composition into its

strong evidence that the enzyme can find functional meaning will be a provocative

the initiation site of ribosomal RNA, and task for the future. Takeichi and Pagano

Suzuki demonstrates the same for silk present evidence implicating proteins on

fibroin messenger RNA, The second ad- cell surfaces in the interaction that these

vance has been developed in collaboration cells can undergo with artificial lipid

with John Gurdon of the ]MRC, Labora- vesicles.

tor>- of ^lolecular Biology, Cambridge, Fambrough and his colleagues con-
England. He has injected 5S RNA genes tinue their detailed and sophisticated
(oS DNA) into the nuclei of frog oocytes, probe of acetylcholine (ACh) receptor
and we have demonstrated that these synthesis and degradation. These recep-
added genes can be transcribed into 5*S tors are glycoproteins and therefore bind
RNA. A new stage of ''genetics by gene certain lectins. This feature is used to
isolation" has begun which involves the show that glycosylation must occur
analysis of important molecules that shortly after receptor synthesis. By
cause genes to be transcribed faithfully in means of a specific inhibitor of this path-
living cells. With this in mind, Wahn way, a dolichol phosphate pathway has
and Reeder continue their analysis of been implicated for glycosylation of re-
the ribosomal DNA chromatin isolated ceptors. Other studies are directed toward
from oocyte nucleoli. tracing receptor proteins from their time

Igor Dawid and his colleagues report of synthesis until their incorporation into

progress on two projects. Some ribosomal the cell membrane.

RNA genes of Drosophila melanog aster The protein bungarotoxin binds specifi-

are known to contain extra stretches of cally to ACh receptors. The probe can

DXA. The function and origin of these be labeled with ^^^I or ^^^I. New receptors

"inserts" is unknown, but the same DNA can be synthesized in the presence of

sequences are located elsewhere in the heavy isotopic precursors so that they are

genome. Ramirez describes the map of distinguished from pre-existing receptor

48 RNA genes throughout the mitochon- molecules. Using a variety of such com-

drial genome of X. laevis. These genes binations, John Gardner and Diana Card

are located all around the circular chro- have carried out detailed studies on the

mosome and on both strands of the DNA. synthesis and degradation of these re-

The medicinal leech provides a simple ceptors.

.system in which to follow the directed Another cell-surface glycoprotein (or

growth of one neuron until it connects group of proteins), acetylcholinesterase,

with another. Muller describes the sprout- is being investigated by Rotundo. Various

ing and directional growth of axons from forms of the enzymatic activity are as-

a crushed neuron toward the distal stump sayed individually. The rates of synthesis

of a neighboring neuron. The target and degradation of these enzyme forms

stump sends detectable transmissions to- as well as their interrelationship will be

DEPARTMENT OF EMBRYOLOGY 9

compared with ACh receptor metabolism, two years at the MRC Unit of Molecular
Fambrough and Pagano have begun a Biology in Cambridge, England, in the
collaborative investigation of the freedom laboratory of Sidney Brenner. Brenner
with which various surface constituents had begun studying the simple nematode
of cells move in the plasma membrane, worm Caenorhabditis elegans. This crea-
They attach fluorescence molecules to the ture has impressive credentials for de-
cell surface and bleach a tiny region of velopmental and genetic research. It has
it with a fine laser beam. The rate at a simple genome and a short life cycle,
which the bleached region is then filled in both of which make it suitable for the
is determined by the fluidity of that por- accumulation of genetic mutants. Dr.
tion of the membrane. The researchers Ward's current interest is probing the
have found that ACh receptors, unlike nature of the surface interactions between
other regions of the membrane, are an- sperm and egg which lead to fertilization,
chored to the surface. He will do this by obtaining mutants

Carbonetto and Muller have discovered that interfere with gamete function.
ACh receptors on the surface of sympa- We acknowledge the indispensable as-
thetic neurons. This was originally found sistance of several agencies and founda-
by bungarotoxin binding, but it has been tions to our research program. Grants
confirmed by electrophysiological means, from the National Institutes of Health
They are beginning studies on the metab- support in part the research programs of
olism and biochemistry of these receptors Brown, Dawid, Pagano, Reeder, and Su-
to determine whether they resemble the zuki. Money from the Whitehall Foun-
well-studied ACh receptors of muscle dation. Inc., has made possible the col-
surfaces, laborative project of Fambrough and
We are especially pleased to welcome Pagano, mentioned earlier in this Intro-
Sam Ward of the Department of Biologi- duction and detailed in the Report. Other
cal Chemistry, Harvard University, as a areas of Fambrough's research have bene-
staff member beginning in the summer of fited from a grant by the Muscular Dys-
1977. He will occupy the laboratory va- trophy Association, Inc. Finally, I can
cated by Jim Ebert. Dr. Ward received report the completion of extensive re-
his Ph.D. in 1971 at the California In- modeling of unused animal quarters into
stitute of Technology, where he studied new laboratory space. This could not
the assembly of bacteriophage by genetic have been done without the generous as-
and biochemical methods. He decided to sistance of a grant from the Max C.
change to eukaryotic systems and spent Fleischmann Foundation.

STUDIES ON SKELETAL MUSCLE AND NEURONAL

PLASMA MEMBRANES

D. M. Fambrough, R. E. Pagano, K. Muller, D. J. Card, J. M. Gardner,

S. Carbonetto, R. Rotundo, and P. N. Devreotes

with the technical assistance of D. Somerville, A. Mabin, W . Duncan, and B. Thomas

Our research interests, broadly defined, nervous system and musculature. ]Much

encompass the regulation of cell-surface of our research has been focused on the

properties during development and the acetylcholine receptors of skeletal muscle

role of cell-surface properties in the in- fibers. These receptors are glycoproteins

teractions between cells. These two inter- embedded in the plasma membranes of

related aspects of developmental biology skeletal muscle fibers; they function as

are manifested most conspicuously and the recognition system for signals from

gloriously in the development of the motor nerves. The number and distribu-

10 CARNEGIE INSTITUTION

tion of acetylcholine receptors in skeletal array of molecular forms by the muscle;

muscle fibers are regulated during devel- it occurs as a cytoplasmic enzyme, a cell

opment and in the continuing interactions surface enzyme, and an enzyme released

between nerve and muscle. Some im- from the muscle fibers into the extra-

portant aspects of this regulation involve cellular milieu. Our attention is being

control of the biosynthesis and degrada- directed more and more to this enzyme

tion of receptor molecules in the muscle because it may help us get at the mecha-

fibers. Thus two of our major research nisms of nerve-muscle interaction during

efforts have been to elucidate the mecha- the formation of neuromuscular connec-

nisms of receptor metabolism, and to de- tions. Another reason for turning toward

termine which mechanisms are most a study of the acetylcholinesterases is

influenced during modulation of receptor that we have developed methods (during

metabolism. We rej^ort here our findings the study of acetylcholine receptor me-

that the lipid intermediate pathway of tabolism) that should permit us to de-

the protein glycosylation (involving termine the relationships of the molecular

dolichol phosphate) operates early in re- forms of the esterase. Finally, compari-

ceptor biosynthesis. The newly synthe- sons between receptor metabolism and

sized receptor molecules, constituting esterase metabolism may shed light upon

about lO^r of all receptors in the system the relation between the biosynthesis of

and already containing significant carbo- plasma membrane glycoproteins and the

hydrate, are located in the Golgi ap- biosynthesis and secretion of secretory

paratus. where they may reside for about glycoproteins.

two hours before transport to the cell sur- Another cell-surface protein that has
face. In the plasma membrane the recep- caught our attention is the binding site
tors are targets for the cells' degradation for a-bungarotoxin on sympathetic gan-
mechanism; this involves interiorization glion cells from the chick embryo. The
of receptor molecules, transport to sec- a-bungarotoxin binding shows character-
ondary lysosomcs, and proteolytic de- istics of binding similar to those of a
struction. Our continuing investigation of cholinergic receptor, but functional tests
receptor degradation has revealed that have been negative. Although we are un-
the normal half-life of a receptor in the certain about the function of this a-
plasma membrane of tissue-cultured em- bungarotoxin binding site, we can still
bryonic skeletal muscle is about 17 hours, focus on this molecule in the studies of
Treatment of receptors with the strongly membrane protein metabolism and its
binding receptor blocking agent a-bun- relation to nerve growth and develop-
garotoxin lengthens the half-life to about ment, thanks to the discovery of the
22 hours. This increased receptor survival stabilizing effect of glutaraldehyde on the
in the plasma membrane helps to explain a-bungarotoxin-binding-site complex, de-
some puzzling observations related to the scribed later in this Report,
existence of "hidden" receptor sites in The interactions between nerve and
skeletal muscle. In the past few years muscle affect not only the metabolism
other investigators have made a strong of the acetylcholine receptors but also the
case for the primary role of muscle ac- receptor distribution on the muscle cell
tivity in modulation of receptor numbers, surface. Within a few days of the forma-
However, we have been unable to demon- tion of functional nerve-muscle contacts,
strate that sort of connection in our ex- most of the extrajunctional receptors are
perimental design. gone, and the remaining receptors are
Another glycoprotein that plays a cru- clustered at high packing density in the
cial role in neuromuscular interactions membrane opposed to the nerve terminal
is the enzyme acetylcholinesterase. This (the postsynaptic surface). The mecha-
enzyme is manufactured in a puzzling nism for clustering of receptors remains

DEPARTMENT OF EMBRYOLOGY 11

a mystery, and little is known about the receptors. As anticipated from earlier

nature of the clustering itself. As part of findings, we found a lag of about 3 hours

an investigation of the interactions of after switching to dense amino acid

lipids and proteins in determining the medium before dense receptors began to

functional properties of plasma mem- appear in the [)lasma membranes. This is

branes, we have been studying these due not to a lag in equilibration of dense

receptor clusters at neuromuscular junc- amino acids with intracellular amino acid

tions. Elsewhere in this Report some pools but rather to a lengthy period of

experiments are summarized which dem- biosynthesis and transport before recep-

onstrate that individual receptor mole- tors are incorporated into plasma mem-

cules at neuromuscular junctions are brane. The evidence that this is so comes

severely restricted in their position. largely from the following observations.

a-Bungaro toxin, which binds readily

and almost irreversibly with acetylcho-

ACh Receptor Biosynthesis jj^^^ receptors located on the cell surface,

P. N. Devreotes and D. M. Famhrough ^^ unable to penetrate the plasma mem-

brane of the muscle cells to interact with

In previous Reports {Year Book 7 4 any receptor sites that might lie along
and Year Book 75) we have described a the biosynthesis-transport pathway. It is
strategy for the direct measurement of found that when muscle cells are grown
biosynthesis of acetylcholine receptors, in the presence of a-bungarotoxin, there
Skeletal muscle is maintained in organ is one set of receptors with which it does
or tissue culture in a medium containing not become complexed. These receptors
amino acids labeled with stable, heavy do, however, become available for «-
isotopes of carbon, hydrogen, and nitro- bungarotoxin binding if membrane struc-
gen. The receptors (and all the other tures are dissociated with non-ionic
proteins made with such amino acids) are detergent solution. Electron microscopic
denser than normal and can be separated evidence for the intracellular location of
from normal receptors by velocity sedi- these receptors will be presented later
mentation. Such dense receptors remain in the report. Our metabolic studies show
responsive to acetylcholine and are syn- that this intracellular set of receptors is
thesized and degraded at rates compar- actually composed of two equal-sized
able to those of normal receptors. Since classes of receptor molecules, containing
the magnitude of the density shift and all together about one-fifth as many re-
the proportion of heavy and light recep- ceptors as are on the cell surface. When
tors can be determined readily, the tech- the cells are switched to a medium con-
nique can be used to measure the kinetics taining dense amino acids, the receptors
of biosynthesis and the degradation of of one class quickly become labeled with
receptor molecules. Its chief advantage is them. After about 3 hours this population
that no purification of receptors is re- of receptors consists entirely of dense
quired. With minor variations the tech- receptor molecules. These receptors are
nique can be used in studies of the therefore precursors to the plasma mem-
metabolism of other proteins, such as brane acetylcholine receptors. The ki-
acetylcholinesterase (see below). netics of labeling of the intracellular

In Year Book 75 we reported on the receptors are illustrated in Fig. 1.

kinetics of appearance of newly synthe- The other class of intracellular acetyl-

sized acetylcholine receptors in the plasma choline receptors, defined by its kinetics

membranes of cultured chick skeletal of labeling, cannot be distinguished from

muscles. We estimated that dense amino the receptors in the plasma membranes,

acids had replaced the normal ones in a Why these sites are not exteriorized, we

ratio of about four to one in the dense do not know. These receptors constitute

12

CARNEGIE INSTITUTION

O

o
a.

Ll_

o

<
en

1

KINETICS OF APPEARANCE
OF % %'%- RECEPTORS
IN PRECURSOR POOL

/
/

A

L4--t-"-'^

■..--i

/F

1

10

TIME (HOURS)

Fig. 1. Kinetics of appearance of ^H, "C, ^^N-receptors in the intracellular pool. [^^]mono-
iodo-a-bungarotoxin-receptor complexes involving the pool receptors were prepared from
cultures incubated for the designated times in medium containing ^H, "C, ^^N amino acids.
These receptors were analyzed by sucrose gradient velocity sedimentation to determine the
fraction of pool receptors which were ^H, "C, ^^N receptors.

a part of the "hidden receptor" popula-
tion referred to in a previous publication
(Devreotes and Fambrough, 1975).

Studies on the Glycosylation of
ACh Receptors

D. Famhrough

Biochemical studies of ACh receptors
from eel and Torpedo electroplax have
demonstrated that these ACh receptors
are glycoproteins. Mammalian ACh re-
ceptors are also glycoproteins, as judged
by their interactions with the lectin
concanavalin A, which binds to terminal
a-mannoside residues of glycoproteins.
When chick .skeletal muscle cells in cul-
ture are treated with concanavalin A, the
receptors can no longer be solubilized
with 1% Triton X-100. Solubilized recep-
tors interact with concanavalin A to form
stable complexes which sediment rapidly
in sucrose gradients. These interactions

are blocked by a-methyl-mannoside, con-
sistent with the specificity of the inter-
action.

Since the ACh receptors undergo such
a long transport period between protein
synthesis and display on the cell surface,
we thought that part of this transport
time might be related to glycosylation of
receptor protein. If this were so, then it
might follow that newly synthesized ACh
receptors would lack the specific terminal
sugar residues that render the mature re-
ceptors susceptible to interaction with
concanavalin A. We found, however, that
all the receptors in the precursor popu-
lation interact with concanavalin A to
form high molecular weight complexes.
Thus, during or very soon after biosyn-
thesis of the receptor polypeptide chains,
a significant amount of carbohydrate is
added to them.

There are two major pathways for the
addition of carbohydrate residues to gly-
coproteins. One is the glycosyl-transferase

DEPARTMENT OF EMBRYOLOGY

13

pathway; the other is the dolichol-
phosphate pathway. In the latter, pre-
assembled chains of carbohydrate resi-
dues are attached to asparagine residues
of the protein. While there are no known
specific inhibitors of the glycosyl-trans-
ferase pathway, a recently discovered
antibiotic, tunicamycin, has been shown
to be a very selective inhibitor of the
dolichol-phosphate pathway.

We obtained a sample of tunicamycin
as a generous gift from Dr. Skip Waechter
(Department of Biochemistry, University
of Maryland School of Medicine) and
used it at a concentration of 1 /xg/ml in
our culture medium in studies of receptor
biosynthesis and incorporation into
plasma membranes and receptor degrada-
tion. At this concentration cultured chick
skeletal muscle fibers survived in good
condition for more than 50 hours (al-
though the fibroblasts in the culture dis-
sociated from the culture plate). The
rate of ACh receptor degradation was
not significantly different from that of

control cultures at 24 hours and only
slightly slower at 50 hours. Tunicamycin
treatment reduced protein synthesis, as
measured by '^H-leucine incorporation
into proteins, by about 50% and blocked
incorporation of ^^C-gluosamine into gly-
coproteins by about 70%. These levels
of inhibition developed over a period of
3-4 hours and were stable thereafter for
another several hours.

In contrast to these partial effects,
tunicamycin completely blocked the bio-
synthesis of ACh receptors within a few
minutes. Incorporation of ACh receptors
into the plasma membrane was not in-
hibited, and thus the intracellular pool
of '^precursor" receptors was transported
to the cell surface and incorporated into
the membrane (Fig. 2). These results
suggest that the carbohydrate moieties
of the receptor include components added
by the dolichol-phosphate pathway and
that these carbohydrate additions are
necessary for the biosynthesis of com-
plete receptor units.

ro

U

I
Surface

1 1
= 96,500 cpm

1

I 1

20

-

\

o/control

-

^

A

A

10

A

+ Tunicamycin

2

1

1 1

1

1 1

2 3 4

Time (hours)

Fig. 2. Incorporation of new ACh receptors into plasma membranes of cultured chick skeletal
muscle (O-O). Effect of tunicamycin (1 /u,g/ml added to medium at time zero) on incorpora-
tion (A-A).

14

CARNEGIE INSTITUTION

Newly Synthesized Receptors
Reside in the Golgi

P. .V. Devreotes and D. Fambrough

There are hundreds of acetylcholine
receptors per square micrometer of plasma
membrane in chick skeletal muscle cells
grown in vitro. These cells also contain
many intracellular acetylcholine receptor
molecules, amounting to about 15-20%
of all the receptors. ]Many of these intra-
cellular receptors are newly synthesized
and have not yet appeared in the plasma
membrane. In the presence of puromycin,
which rapidly halts protein synthesis, the
intracellular pool of newly synthesized

receptors decreases approximately line-
arly to near zero in about 3-4 hours as
an equal number of receptors appears in
the surface membrane. Thus, in the
present study we were attempting to de-
termine the subcellular location of a set
of receptor sites which (1) were unavail-
able for interaction with extracellular
receptor ligands, (2) were exposed to
such ligands by techniques that render
lipid bilayers permeable to large mole-
cules, (3) constituted about 10-15% of
all the receptor sites, and (4) decreased
steadily in number when protein syn-
thesis was inhibited until few such sites
were present after 3-4 hours.

b

I.U

^

1

1 1 1

1

E

\

o

E

0.8

•

^^

\

O

\^

o

o

0)

0.6

X

or

^^

o

*—

" —

X^

o

c

.^-— i.

•

0)

0.4

-

•

c

**-

o

c

0.2

—

'

—

o

'.«_

o

o

^

Ul

1

1 i 1

1

1

2 3 4

Time (hr) in Puromycin

Fig. 3. Decline in saponin-revealed ACh receptors in cultured chick skeletal muscle after
inhibition of protein synthesis by puromycin. Cultures were treated with unlabeled a-bun-
garotoxin to block all surface ACh receptors, and puromycin was added to the medium of
subsets of cultures at different times. Then all cultures were rinsed briefly to remove medium
and were fixed for 1 hr at room temperature in fresh formaldehyde. After fixation the cultures
were treated with 0.5% saponin solution for 3 min and rinsed again. Then the cultures were
incubated at room temperature in medium containing 0.03 to 0.05 Mg/ml a-[^T]bungarotoxin
for 90 min, rin.sed 30 min with large volumes of Hanks' balanced salt solution, and the bound
radioactivity counted. Specific binding was defined as binding not blocked by preincubation
of cultures for 15 min with 1.5 X 10~* M d-tubocurarine and inclusion of this concentration
of d-tubocurarine in the medium containing a-[^T]bungarotoxin. Open circles represent
averaged data from two to four experiments involving four to eight culture dishes for each
point. Closed circles represent data from single experiments with four to eight culture dishes
for each point.

DEPARTMENT OF EMBRYOLOGY 16

We found such a population of receptor also produced 50% inhibition. The kinet-

sites in chick muscle cells fixed in fresh ics of a-[i^^I]bungarotoxin binding t^ the

formaldehyde by a modification of the surface and to the saponin-revealed sites

methods of Daniels and Vogel (Nature were indistinguishable. In each case the

254, 339, 1975). These sites were una vail- binding curve approximated a first-order

able for interaction with extracellular exponential with a half-time of 25 min-

a-bungarotoxin for 24 hours prior to utes when 5 nM a-bungarotoxin was

fixation. After fixation, the intracellular bound at 25°C.

pool of newly synthesized receptors could Essentially all (>95%) of the binding

be partly or completely exposed to a- sites revealed by saponin treatment were

bungarotoxin by first treating the fixed intracellular, as determined by electron

cells with (1) cycles of freezing and microscope autoradiography. Particularly

thawing, or (2) Triton X-100 or sodium striking was the frequent occurrence of

cholate, by brief extraction or (3) 0.5% silver grains directly over the Golgi ap-

saponin, a lipophilic antimicrobial agent, paratus (Fig. 4). In cultured chick skele-

Of these procedures, the saponin treat- tal muscle cells the Golgi apparatuses

ment proved most useful because it mini- almost invariably sit adjacent to nuclei,

mally altered the ultrastructure of the In our autoradiographs of control cul-

muscle cells and quantitatively revealed tures 17% of silver grains attributable

the intracellular receptor pool within a to specific binding of labeled a-bungaro-

few minutes. The specific receptor sites toxin were directly over easily identifi-

revealed by saponin constituted about able Golgi. Another 23% of such silver

15-20% of all the sites. When the cul- grains were located over membranous

tured chick muscle cells were treated with elements in the cytoplasm adjacent to a

puromycin and subsequently fixed, the nucleus, but they could not unequivocably

number of specific intracellular receptor be identified as Golgi membranes. The

sites revealed by saponin treatment was remaining 60% of the silver grains asso-

decreased as expected (Fig. 3) . The kinet- ciated with specific binding occurred pre-

ics of decrease are very similar to those dominantly over membranous cytoplas-

obtained when the intracellular sites were mic structures that could not readily be

measured in detergent extracts obtained identified. There is very little rough

from living cells. endoplasmic reticulum in these cells, and

The number of acetylcholine receptors silver grains over rough endoplasmic

in these experiments was considered to be reticulum have not been seen so far in

the number of specific binding sites for this study.

^^^I-labeled a-bungarotoxin. Specific bind- If the specific a-bungarotoxin binding
ing sites were defined as sites that could sites associated with the Golgi apparatus
be completely protected from a-bungaro- represented a portion of the pool of intra-
toxin binding by 10~^ M d-tubocurarine. cellular receptors that were in the path-
The inhibition of a- [^^^I] bungarotoxin way of membrane biogenesis, then there
binding on the specific sites was studied should be fewer of those sites when pro-
as a function of d-tubocurarine concen- tein synthesis is inhibited while recently
tration and acetylcholine concentration, synthesized receptors are being incor-
Inhibition curves were similar for the porated into the plasma membrane. In
surface sites and the saponin-revealed fact, the time course of loss of receptor
sites, a- [^2^1] Bungarotoxin binding was sites from the Golgi should reflect the
inhibited 50% by about 3-5 X 10""^ M chronological position of the Golgi in the
d-tubocurarine, which is similar to the pathway by which receptors proceed
concentration needed to inhibit 50% of from their sites of synthesis to their in-
the binding with living cells. A concen- sertion points in the plasma membrane,
tration of 1-4 X 10~^ M acetylcholine We scored the location of silver grains

16

CARNEGIE INSTITUTION

Fig. 4. Electron microscope autoradiographs illustrating the association of ACh receptors with
the Golgi apparatus, as revealed by the binding of a-[^""'I]bungarotoxin to fixed, saponin-
treated chick skeletal muscle cells.

in autoradiographs of cells fixed at differ-
ent times after inhibition of protein syn-
thesis by puromycin. Figure 5 shows that
the specific association of receptors with
the Golgi begins to decrease soon after
the inhibition of protein synthesis and
is no longer evident 2-3 hours later. Thus

newly synthesized acetylcholine receptors
are located in the Golgi apparatus for
about half of their intracellular residence
time, mostly during the first half.

Further studies along the lines de-
scribed here should (1) establish more
precisely the time course of association

DEPARTMENT OF EMBRYOLOGY

17

Time

2
(hrs)

in

3 4

Puromycin

Fig. 5. Fraction of nowly F-ynthcsizod ACh
receptors in tl'ie Golgi apparatus after inhibition
of protein synthesis by puromycin. Experiments
were performed as described in the legend to
Fig. 3, except that after the operations de-
scribed there, the cultures were fixed overnight
in 2% glutaraldehyde and then osmicated,
stained with aqueous uranyl acetate, and de-
hydrated and embedded in Epon plastic. 1000
A sections were cut and processed for electron
microscope autoradiography. All the silver
grains over many cells were counted and the
underlying structures classified. Silver graim?
were scored as related to the Golgi apparatus
only if they were directly over membranes of
the Golgi and did not overlie any portion of
nucleus or nuclear membrane. In the absence of
puromycin, 8.5% of the silver grains were over
the Golgi and for cultures 5 hr in puromycin
about 2% of silver grains were so located. In
these experiments the intracellular pool of
newly synthesized receptors represented about
25% of the silver grains in the absence of puro-
mycin. It was assumed (based upon much pre-
vious data, including that of Fig. 3) that in the
presence of puromycin the intracellular pool of
newly synthesized ACh receptors declined
linearly to zero over a period of 3.5 hr. The
fraction of total silver grains over Golgi minus
0.02 was divided by the fraction of silver grains
representing the new receptor pool to obtain
the points plotted here. This representation
probably depicts the time course of loss of
receptors from the Golgi fairly accurately, but
it certainly underestimates the fraction of re-
ceptors associated with the Golgi. This is be-
cause silver grains not directly overlying Golgi
are not included, and silver grains overlying
perinuclear membranes which could not be
unequivocally identified as Golgi were likewise
excluded.

of newly synthesized ACh receptors with
the Golgi, (2) locate the receptor sites
more precisely in certain elements of the
Golgi and establish the polarity of the
receptor molecules with respect to intra-
cellular membrane systems, (3) explore
the possible association of new receptor
sites with rough endoplasmic reticulum,
and (4) help to determine the sites of
action of various inhibitors of ACh re-
ceptor biogenesis (such as X537A, tunica-
mycin, 2,4-dinitrophenol, 2-deoxyglucose,
and D-glucosamine) .

Regulation of Acetylcholine Receptor
Metabolism by Electrical Stimulation

D.J. Card

Acetylcholine receptor metabolism in
cultured myotubes and denervated adult
muscle can be regulated by the amount
of contractile activity of the muscle cell.
If muscle contractions are abolished by
the application of tetrodotoxin or lido-
caine or by denervation, the number of
extrajunctional acetylcholine receptors in
the surface membrane increases. Contrac-

18 CARNEGIE INSTITUTION

tions elicited by decreased extracellular tion but an acceleration of the degrada-

potassium (in mouse myotubes), by di- tion rate by a factor of more than 20.

rect electrical stimulation or by reinner- Alternatively, electrical stimulation could

ration of a muscle cause a dramatic cause a decrease in the number of recep-

decrease in the number of extrajunctional tors by affecting both incorporation and

receptors. Most previous studies have degradation rates simultaneously.

measured acetylcholine sensitivity or the

total number of acetvlcholine receptors t-. ■, .-

, , r ' ^T^ • X i 1 Degradation
on the muscle surface. \\ e are mterested

in which aspects? of acetylcholine receptor The degradation rate of extrajunc-

metabolism (synthesis and incorporation tional acetylcholine receptors can be in-

into the plasma membrane, the subse- f erred from measurements of the rate at

quent internalization of the receptors, or which ^^^I-labeled a-bungarotoxin bound

their ultimate degradation) are most to these receptors is degraded. Concom-

affected by muscle activity. Therefore, itantly, radiolabeled degradation prod-

we have modified techniques already in ucts are released from the muscle fibers.

use in this laboratory in order to study Degradation rate in adult denervated

the effects of electrical stimulation on muscles in organ culture was measured

denervated adult tissue. Reports on the as follows. First, the muscle was exposed

effect of stimulation in primary tissue to a saturating dose of a-[^^^I]bungaro-

culture and preliminary data on dener- toxin. Then the residual unbound toxin

vated adult muscle were presented last was rinsed away. The muscle was tied

year. with 5-0 silk around its tendons onto

Following denervation, the num^ber of posts in a Plexiglas chamber containing

acetylcholine receptors in adult muscle Trowell medium and 0.1% bovine serum

increases, and the acetylcholine sensi- albumin. In the stimulation experiments

tivity in rat soleus muscle rises to a muscles were positioned over platinum

plateau level of about 250 mV/nC. Elec- wire electrodes with one end of the muscle

trical stimulation (given at 100 Hz for affixed to a Grass force transducer in

1 second, every 100 seconds) causes a order to monitor muscle tension. Contrac-

fall in acetylcholine sensitivity to the tile tension in all muscles was noted at

upper limits of normal (about 1 mV/nC) the beginning and end of the incubation

in 4 days (Lomo and Westgaard, J. period. The muscles were incubated in

Physiol ^252, 603, 1975). These changes a humidified 95% O2, 5% CO2 atmos-

in acetylcholine sensitivity are compara- phere at 37°C. At regular intervals dur-

ble to a change in receptor number from ing incubation the Trowell medium sur-

100 receptors per unit area to 3 receptors rounding the muscle was removed and

per unit area in 100 hours. saved for analysis, and fresh warm

The total number of acetylcholine re- medium was replaced in the chamber,

ceptors on the surface membrane at any The degradation medium was analyzed

time is the result of a balance of the for total amount of radioactivity and for

synthesis-incorporation rate and the the amount of radioactive label associ-

degradation or removal rate. Therefore ated with specific molecules, namely

the dramatic decrease in the number of ^^^I-tyrosine. As a result of acetylcholine

surface receptors within 100 hours with receptor degradation, ^^^I-tyrosine ac-

100-Hz electrical stimulation could be cumulates in the culture medium. The

the result of (1) a complete inhibition rate of appearance of ^^^I-tyrosine paral-

of the production of new acetylcholine lels the loss of radioactivity from the

receptors, with no change in degradation muscle surface (Fig. 6).

rate, or (2) at the other extreme, no Degradation rate for the rat extensor

change in receptor synthesis-incorpora- digitorum longus (a fast-twitch muscle

DEPARTMENT OF EMBRYOLOGY

19

20 30

Hours

with suprathre.sholr] stimuli. The succes-
sive stimuli in each train were of alter-
nate polarity to minimize the possibility
of polarization of the stimulating elec-
trodes. There was little change in degra-
dation rate as a result of stimulation. In
some cases there was a slight reduction
in degradation rate, e.g., the half-life for
receptor degradation in EDL muscle in-
creased from 22 to 23 hours.

Thus the same pattern of electrical
stimulation which has been shown to
promote rapid decline in extrajunctional
acetylcholine chemosensitivity in vivo
results in a small increase in receptor
half-life measured in vitro. Since acetyl-
choline receptors decrease during electri-
cal stimulation (Lomo and Westgaard,
1975) without any acceleration of degra-
dation (Hogan, Marshall, and Hall,
Nature 261, 328, 1976; Shainberg and
Burnstein, Nature 264, 368, 1976), it fol-
lows that electrical stimulation must
cause a very large decrease in either the
biosynthesis of new receptors or their rate
of incorporation into plasma membranes.

Fig. 6. Degradation of a-[^^^I]bungarotoxin-
receptor complexes in 8-day denervated rat
extensor digitorum longus muscle. The fraction
of radiolabeled receptors remaining on the
muscle is plotted at different incubation times.
The half-life is approximately 20 hr.

relying predominantly on anaerobic
metabolism) was 2.3% per hour, a half-
life in the membrane of approximately
22 hours. Acetylcholine receptors in rat
soleus (a slow-twitch muscle with pri-
marily oxidative metabolism) were de-
graded at 2.7% per hour, a half-life of
18 hours. Degradation rate in these mus-
cles was independent of length of dener-
vation (from 5 to 11 days). Degradation
rate was decreased by reducing the tem-
perature to 25°C and adding cyclohexi-
mide to the culture medium.

To measure the effect of electrical
stimulation on degradation rate, the
muscles were stimulated for 5-12 hours
at 100 Hz for 1 second every 80 seconds

Incorporation

Incorporation of acetylcholine recep-
tors into muscle plasma membranes was
measured by the appearance of new oc-
bungarotoxin binding sites at various
times after the existing receptors were
blocked with unlabeled a-bungarotoxin.
The experimental protocol was as follows.
Five-day denervated rat muscles were
put in Plexiglas chambers. The muscles
were tied in place at in situ resting length
and barely covered with Trowell medium,
0.1% bovine serum albumin, 5 X 10~^
g/ml gentamicin, 2 X 10~^ g/ml fungi-
zone, 200 units/ml penicillin, 5 X 10~^
g/ml streptomycin. The muscles were
segregated into four experimental groups
to determine the total number of: (1)
acetylcholine receptors, (2) receptors
available for a- [^^^I] bungarotoxin bind-
ing immediately after saturation of the
surface receptors with unlabeled a-bunga-
rotoxin, (3) receptors incorporated into

20

CARNEGIE INSTITUTION

the plasma membrane during incubation
after treatment with unh\beled a-bunga-
rotoxin. and (4) receptors incorporated
during electrical stimulation at 100 Hz
for 1 second every SO seconds.

The muscles were incubated for various
lengths of time in normal Trowell me-
dium. Then unlabeled a-bungarotoxin
(1.68 fig ml) was added to some of the
muscles (groups 3 and 4) in order to
saturate the receptors. Residual a-bunga-
rotoxin was rinsed away, and the muscles
were incubated for various times to await
the appearance of new receptors. The
number of receptors on all muscles was
determined by placing the muscles in a
solution of a-[^-^I]bungarotoxin at the
end of the experiment. Unbound toxin
was rinsed away. Then the muscles were
homogenized and the toxin-receptor com-
plexes extracted from a membrane pellet
(10.000^ for 30 min) with 2.57c cholate,
10 mM tris. The number of a-bungaro-
toxin-receptor complexes per milligram
of muscle tissue was determined from the
amount of radioactive material moving
as a lOS peak by velocity sedimentation
in a 5-ml 5-20% sucrose gradient.

Experiments were also carried out
where ^H. ^^C, ^^N-labeled amino acids
were substituted for ^H, ^^C, ^^N amino
acids in the Trowell medium. In these
experiments, if the receptors that appear
on the muscle surface are indeed the re-
sult of de novo synthesis, they will be
composed of the 2H, i^C, i5;N-labeled
components. Therefore, the new receptors
will exhibit a density shift that can be
detected by velocity sedimentation. The
protocol in these experiments was similar
to the former incorporation measure-
ments. The muscles were put into three
experimental groups. The first group was
used to study the incorporation of new
receptors. Existing receptors were blocked
with unlabeled a-bungarotoxin, and the
emergence of newly synthesized ''heavy"
receptors was measured. In the second
group, the existing receptors were blocked
with unlabeled a-bungarotoxin and the
incorporation of new ''heavy" receptors

was determined during electrical stimula-
tion of the muscle at 100 Hz for 1 second
every 80 seconds. IMuscles of these two
experimental groups were placed in me-
dium containing -H, ^^"^C, ^^N amino acids
while the surface acetylcholine receptors
were blocked by unlabeled a-bungaro-
toxin, so that any precursor receptor
molecules would be made from the
"heavy" amino acids. Then after the un-
bound a-bungarotoxin was rinsed away,
the muscles were again placed in the ^H,
^^C, ^^N-labeled medium for various
lengths of time. The third group was used
to determine the total number of recep-
tors on muscles not treated with un-
labeled a-bungarotoxin. Muscles of this
group were cultured in medium contain-
ing ^H, ^^C, ^"^N amino acids, and the
number of receptors was determined at
the end of the incubation period. The
third group of muscles is very important,
since all incorporation data are expressed
as percent of total number of receptors.
The number of receptors on muscles of
all groups was determined by saturating
them with a-[^^^I]bungarotoxin at the
end of the experiment. Toxin-receptor
complexes were isolated as previously de-
scribed. The amount of toxin-receptor
complexes was determined from the
amount of radioactive material in the
toxin-receptor peak after velocity sedi-
mentation in sucrose-deuterium oxide
gradients.

In order to provide a standard of nor-
mal toxin, ^H, ^^C, ^*N-labeled receptor
complexes in each centrifuge tube, toxin-
receptor complexes were prepared from a
fourth group of denervated muscles,
freshly dissected and then saturated with
a- [^"^^I] bungarotoxin in medium contain-
ing ^H, ^^C, ^^N-labeled amino acids.
After the toxin-receptor complexes were
extracted, a portion of the a-[^^^I]-
bungarotoxin-receptor complexes was
added to each sucrose gradient tube, in
addition to a portion of extract from an
experimental muscle. Using the a- [^^^I] -
bungarotoxin ^H, ^^C, ^*N-receptor com-
plex peak of the sucrose gradient as a

DEPARTMENT OF EMBRYOLOGY

21

template, the proportions of a-[^^^I]-
bungarotoxin ^H, ^-"^C, ^^N-receptor com-
plexes and a-[i2^I]bungarotoxin ^H,
i2Q^ i4;fy^_j.g(>gp^Qj. complexes on each

gradient were estimated (Fig. 7). Theo-
retical superposition of ''heavy" and
''light" toxin-receptor complex peaks of
different proportions was carried out with
the aid of a DEC PDP-8 computer. The
best fit was obtained by visual inspection.
Both techniques of measuring incorpora-
tion rate of receptors into extra junctional
membranes gave similar results.

We discovered that the amount of

3000-

2000-

1000-

10 20 30

Fraction number

40

Fig. 7. A typical sucrose deuterium oxide
gradient of toxin-receptor complexes from 5-
day denervated extensor digitorum longus
muscle that was stimulated in vitro for 36 hr.
Receptors were allowed to incorporate for 9^/2
hr in ^H, ^^C, ^^N-labeled medium. Incorpora-
tion rate was 1.1% per hour. Toxin-receptor
complexes extracted from muscle incubated in
'H, ""C, "N-labeled medium (treated with a-

[^^I]bungarotoxin) are less dense ( ) than

42% of the toxin receptor complexes incubated
in 'B, "C, ^^N-labeled medium (a-[^^^]bunga-

rotoxin) ( ). The proportion of a-[^T]

bungarotoxin ^H, "C, ^^N-receptor complexes
was estimated as described in the text.

vigorous muscle contraction was very
important in producing any changes in
acetylcholine receptor metabolism. There-
fore, during the stimulation period the
voltage of the stimulator was adjusted
so that it was twice that needed to elicit
a muscle twitch. This voltage was usually
increased during the experiment in order
to maintain the level of muscle contrac-
tion. Between 12 and 26 hours of stimula-
tion at 100 Hz every 80 seconds, the
incorporation rate of stimulated muscles
decreased to 30% of the incorporation
rate in unstimulated control muscles. A
similar decrease occurred in muscles stim-
ulated for 42 and 69 hours with the same
pattern of 100-Hz stimuli. We are now
investigating stimulation times between
12 and 26 hours more closely to deter-
mine the kinetics of decline in rate of
receptor incorporation.

Other incorporation data indicate that
muscles stimulated at levels that elicit a
visible twitch but not a maximal con-
traction, at 100 Hz intermittently for up
to five days in vitro incorporated acetyl-
choline receptors into their membranes
at a rate that was not significantly differ-
ent from muscles stimulated only 4 hours
(see Fig. 8). The slope of the least-
squares fit indicates a very slow^ decrease
in receptor incorporation rate over 120
hours in culture. This is true for dener-
vated rat extensor digitorum longus as
well as denervated soleus muscles.

The only difference between these and
the previous incorporation experiments
is the amount of voltage applied across
the muscle. The muscle must be contract-
ing vigorously in order to induce changes
in acetylcholine receptor numbers. Thus
it is apparent that the number of supra-
threshold-activated muscle fibers is very
important in the regulation of acetyl-
choline receptors. Subthreshold activa-
tion is not adequate to produce changes
in receptor metabolism.

Our results suggest that suprathreshold
activation of denervated muscle by inter-
mittent 100-Hz electrical stimulation
leads to a dramatic decrease in acetyl-

22

CARNEGIE INSTITUTION

0)

3

E

CO

1.8

I 1

I

•
o

1111

EDL muscles in 'h.'^C.'^N medium

EDL muscles in^H.'^C.'^N medium

14

-

• •

X

Soleus muscles in ^H.'^.'^N medium'

1.0

-

•

X

:

o -

f- §

: °

o

0.6

-

2

1 1

1

1

1 1 1 1

10

20 30 40 50 60 70

Duration of stimulation (hrs.)

80

90

100

Fig. S. Comparison of incorporation rates in stimulated and control muscles after different
durations of stimulation (incorporation rate in stimulated muscle divided by incorporation
rate in control muscle). Linear regression analysis reveals that even after a short time, the
stimulated muscle has a shghtly smaller incorporation rate than the control muscle, perhaps
because of the stress of stimulation.

choline receptor incorporation rate be-
tween 12 and 26 hours of stimulation. A
closer examination of the onset of the
stimulation-induced decrease in receptor
synthesis will undoubtedly tell us more
about the precise control point (s) in-
volved in regulation of acetylcholine
receptor metabolism.

Degr.\datiox of Acetylcholine
Receptors

John M. Gardner

Earlier measurements of ACh receptor
turnover were based on the time-depend-
ent degradation of mono-iodo-a-[^^^I]-
bungarotoxin bound to ACh receptors. It
was reasoned that the toxin-receptor in-
teraction was so strong that the complex
wouhl be degraded as a unit by the cells,
anrl a variety of tests indicated that the
binding of a-bungarotoxin did not greatly
alter the rate of receptor turnover. How-
ever, it still seemed desirable to obtain
a direct measure of the rate of receptor
turnover. This was accomplished, using
the density shift technique described
above anrj in previous publications (Year
Book 74, Year Book 75).

A variety of strategies based on the

density shift technique were employed
in this study. In this report we have
assigned a short descriptive name for
each experimental design and have illus-
trated the designs in Fig. 9. Some addi-
tional details of experimental procedure
are given in the figure legend.

The first two kinds of experiments were
performed to determine whether cells
grown in the presence of ^H, ^^C, ^^N
amino acids turned over ACh receptors
at the same rate as cells grown in normal
medium and whether receptors labeled
with the dense amino acids turned over
at the same rate as those synthesized
from normal light amino acids. The
''density-shift^toxin-dcgradation" experi-
ment is shown in Fig. 9A, and the '^den-
sity-washoutr-toxin-degradation" experi-
ment in Fig. 9B. In the former case,
myogenic cells were plated in normal
medium and allowed to differentiate and
synthesize a set of normal ACh receptors.
These receptors were labeled with «-
[^^•''Ilbungarotoxin, and the cultures
switched to medium containing dense
amino acids. In the latter case myogenic
cells were plated and grown in medium
containing dense amino acids so that all
their ACh receptors were density-shifted.
These receptors were labeled with «-

DEPARTMENT OF EMBRYOLOGY

23

FVe-experimental

Experimental

fwm

Fig. 9. Experimental strategies employed in
ACh receptor turnover studies. (A) Density-
shift-toxin-degradation; (B) density-washout-
toxin-degradation; (C) density-shift; (D)
density-washout; (E) pulse-chase; (F) two-
population experiment. The solid bars indicate
the time intervals during which cells were in
normal medium containing ^H, ^^C, ^^N amino
acids. The cross-hatched bars indicate the
growth of cells in "dense" medium (containing
^H, "C, ^N amino acids). The large soHd arrows
represent the application of [^^T]mono-iodo-a-
bungarotoxin to all the cultures in the experi-
ment for just enough time to saturate the
receptors on the cells; the unbound a-bungaro-
toxin was then rinsed away and the cultures
returned to the incubator. The small solid ar-
rows indicate the time points at which small
subsets of cultures were incubated with a-[^^T]
bungarotoxin, rinsed and then the receptors
were extracted and analyzed by velocity sedi-
mentation. The small open arrows indicate the
same procedure except that a-[^^I]bungarotoxin
was used.

[^2^1] bungarotoxin, and the cultures
switched back to normal culture medium.
In both cases the loss of radioactivity
from the cells was monitored for at least
48 hours. In the density-shift-toxin-
degradation experiment (Fig. 9A) a-
bungarotoxin-receptor complexes were
extracted from the cultures at each time-
point and analyzed by velocity sedimen-
tation, using a a- [^^^I] bungarotoxin-

light-receptor marker in each gradient.
In all cases, 100% of the a-V'^n']\>\\r\^h-
rotoxin-receptor complexes were 10»S,
indicating that the labeled a-bungaro-
toxin never dissociated from the light re-
ceptors and rebound to heavy receptors.
Similarly, the a-[^^^I]})ungarotoxin-re-
ceptor complexes in the density-washoutr-
toxin-degradation experiments (Fig. 9B)
were 100% density-shifted throughout
the experiment. The results of the two
types of experiments are shown in Fig.
10. In both cases the loss of a-[^^^I]-
bungarotoxin-receptor complexes fol-
lowed first-order kinetics with a half-time
of 22-24 hours. These values are in good
agreement with the established 22-hour
half-time of toxin receptor complexes
under normal culture conditions. Thus
the turnover of acetylcholine receptors
to which a-bungarotoxin is bound takes
place at a rate not significantly influ-
enced either by a culture medium con-
taining ^H, ^^C, ^^N amino acids or by a
high degree of substitution of these amino
acids in the polypeptide chains of the
ACh receptors.

To measure the turnover rate of the
native ACh receptors without bound «-
bungarotoxin, experimental designs C and
D (Fig. 9) were used. These are termed
"density-shift" and ''density-washout"
experiments, respectively. Chick muscle
cultures were prepared using normal
medium or medium with dense amino
acids. When a large population of recep-
tors was present in the cultures, the
medium was switched from normal to
dense or from dense to normal, and
cultures were maintained in the new
medium for the course of the experiment.
At each time point the ACh receptors in
a small subset of the cultures were labeled
by brief exposure to a- [^^^I] bungaro-
toxin, and the labeled receptors were
solubilized in non-ionic detergent and
analyzed for normal and density-shifted
receptors by velocity sedimentation in
sucrose gradients. The total number of
normal receptors remaining on the cells
switched to dense medium and the total

24

CARNEGIE INSTITUTION

C3»

c

o

E
a>
(T

if)

O

Q.

a>
o

0)

cr

JO

o

o

c

CD

c

O

o
o

^H,'^C, '^N - Receptors

0.1

10 20 30 40

Time ( hours )

50

60

70

Fig. 10. Degradation of a-[^"T]bungarotoxin-receptor complexes; normal and dense recep-
tors compared. (O-O) toxin-dense receptor complexes in normal medium; (A- A) toxin-
normal receptor complexes in dense medium.

number of den.sity-shifted receptors re-
maining on the cells switched to normal
medium were plotted as a function of
time. These curves could readily be ap-
proximated by first-order exponentials
with half-decay times of 15-17 hours
(Fig. 111. The time course for the dis-
appearance of normal and of density-
shifted receptors was the same. This
was confirmed for rjonsity-shifted and
normal receptors on the same set of mus-
cle cultures by experimental design E

(Fig. 9) , which is a hybrid of designs
C and D. In this experiment the half-
times for degradation of normal and of
density-shifted receptors were both 17
hours.

The final experimental design (Fig.
9F) was termed the ''two-population"
experiment. In this experimental design,
degrarlation rates of receptors with and
without bound a-bungarotoxin were meas-
ured on the same cells. This experiment
is thus internally controlled, so a differ-

DEPARTMENT OF EMBRYOLOGY

25

o»

o

£

<n
o

Q.
<D
O
0)

q:
o

c
■><

o

o

o
o»

c

CD

o
o

^H,'^C, '^N -Receptors

10 20 30 40

Time ( hours )

50

60

70

Fig. 11. Degradation of normal and density-shifted receptors without bound a-bungarotoxin.
(O-O) dense receptors ; (A-A) normal receptors.

ence in turnover rates cannot be ascribed
to variations between experiments. Un-
fortunately, this experimental design is
the most difficult. The results suggest
that when a-bungarotoxin is bound to
ACh receptors, their degradation rate is
somewhat slower than that of normal
receptors. Myogenic cells were plated in
normal medium and differentiated to pro-
duce normal receptors. These receptors
were labeled with ^-[^^^Ijbungarotoxin,
and the cultures were returned to normal
medium for 18 hours, during which time

a new set of unlabeled, normal ACh
receptors was synthesized. Then the cul-
tures were switched to medium contain-
ing dense amino acids. The disappearance
of the normal receptors during culture in
the dense medium was monitored by
saturating the free receptors in small
subsets of cultures at different times,
using a- [^"^^I] bungarotoxin and then ana-
lyzing the fraction of a-[^2^I]bungaro-
toxin-receptor complexes and a-[^^^I]-
bungarotoxin-light-receptor complexes
after velocity sedimentation of the deter-

26

CARNEGIE INSTITUTION

gent extracted receptors. The data from
two experiments are plotted in Fig. 12.
In both experiments the loss of free,
normal ACh receptors from the cultures
followed first-order kinetics with a half-
decay time of 17-1 7.5 hours, Thea-[^^^I]-
bungarotoxin-normal receptor complexes
were lost at rates of 19 and 22 hours.
Thus receptors to which a-bungarotoxin
is bound are degraded a little more slowly
than free receptors.

There are many possible explanations

for the effect of bound a-bungarotoxin
on receptor turnover. For example, the
a-bungarotoxin may (1) provide a slight
protection from proteolytic attack by
sterically hindering the access of pro-
teases to some portions of the receptor,
or (2) hold the receptor in a conformation
more metabolically stable than other con-
formations, or (3) block the spontaneous
denaturation of receptor molecules that
might contribute to the overall degrada-
tion rate, or (4) change the frequency

1.0
0.9

0.8

0.7

0.6-
Q5-

c

o 0.4

E

I 0.3

o
o

0.2-

0.1

\ 1 1 1 1 1 1

-V

-

X\

"\

^

- \^

\. \^ "-Jl-o-Bungarotoxin -

-

V\ A ^ Complexes (A,A)

W "^ °

-

\ V \

V \

\^ \

\^ ■-

\\ '^

\^ N

\ \ \

\ \ "^

A\ -

■"

/ V\ A\

(•,o) 'H.'^C'^N-Receptors-^ a\^\

(free) \ '^

\«\

\ \

\ \

1 1 1 1 1 1

10 20 30 40

Time ( hours )

50

60

70

Fig. 12. Degradation of normal receptors with (A, A) and without (•, O) bound a-bungaro-
toxin in the pre.«ence of den.se medium : the two-population experiment. Data from two experi-
ments are graphed; the solid symbols refer to one experiment, the open symbols to the other.

DEPARTMENT OF EMBRYOLOGY 27

with which receptors occur in areas of 37°C. We surmised that this difference

membrane internalized by the cell for might be related to the "hidden sites"

transport to lysosomes. being available for slow interaction with

Two years ago we published the results «-bungarotoxin at high temperature but

of studies which indicated that a-bunga- not at low temperature (Year Book 74).

rotoxin did not have much effect on the Our new data on the degradation of

rate of receptor degradation (Devreotes receptors with and without bound «-

and Fambrough, J. Cell Biol. 65, 335, bungarotoxin provide a less mysterious

1975). In those experiments we blocked explanation for much of the data related

receptor biosynthesis and monitored the to ''hidden" receptors. To measure the

degradation of a-bungarotoxin bound to slow binding of a-bungarotoxin to recep-

the acetylcholine receptors or, in the tors of the ''hidden" class at 37°C, the

absence of a-bungarotoxin, the disappear- experiment must be designed to take into

ance of the receptors. We found then that account the fact that muscle continues

the equivalent of 847 receptors per unit to incorporate new receptors all the time,

area of cell surface were degraded each Therefore, we set up experiments to

hour in the absence of a-bungarotoxin, measure the binding kinetics by adding

and 810 per hour when it was present, labeled a-bungarotoxin to sets of cultures

This kind of experiment was technically at different times and then simultane-

difficult, and we concluded that the dif- ously washing the unbound toxin from

ference in rates lay within experimental all cultures at the end of the experiments

error and thus that any perturbation of and measuring the counts bound. It was

receptor degradation by a-bungarotoxin assumed that at the end of the experi-

must be small. As indicated above, our ment all cultures would contain the same

density-shift experiments show that the total number of receptors. The new data

rate of receptor degradation is slightly on receptor degradation demonstrate that

diminished in the presence of bound a- this assumption is not valid. Since a-

bungarotoxin. How, then, should this bungarotoxin stabilized receptors slightly

measured effect of a-bungarotoxin mani- against degradation, the cultures treated

fest itself in the indirect test? The answer for long times in media containing a-

is that if the unperturbed degradation of bungarotoxin actually have extra recep-

receptors destroyed 847 receptors per tors — those spared from degradation by

unit area of cell surface per hour, then the bound toxin. This sparing of recep-

the degradation of receptors with a- tors will have an effect on the total

bungarotoxin bound to them should occur receptor number, which can be quantita-

at a rate of 808 receptors per unit area tively predicted if one knows the rates of

per hour. Thus there is no conflict be- receptor production and degradation dur-

tween the density-shift data and the less ing the binding experiment. For example,

precise older data with regard to the if the cultures have a steady-state level

effect of a-bungarotoxin on receptor of receptors in the absence of a-bungaro-

degradation. toxin, then adding a-bungarotoxin will

Several different kinds of experiments result in an increase of about 15% in

have suggested that some acetylcholine total receptor number over 24 hours. We

receptors besides the newly synthesized have recently done several complicated

ones are also not readily available for experiments to evaluate this point. These

interaction with a-bungarotoxin in the experiments, which cannot be presented

culture medium. We have termed such in detail here, confirm that a-bungaro-

sites "hidden receptors." One interesting toxin stabilization of receptors against

aspect of the interaction of a-bungaro- degradation can approximately account

toxin with cultured muscle cells is that for the slowly saturable "hidden" sites

fewer sites are labeled at 4°C than at we postulated earlier. There remains

28

CARNEGIE INSTITUTION

another set of "hidden" receptors: those
which cannot be hibeled by extracelhilar
a-biingarotoxin and yet are not recently
synthesized. This popuhition frequently
represents about 10-1 5 <^r of the receptors
in cultured chick skeletal muscle cells.

Positional Stability of Acetylcholine

Receptors at the Neuromuscular

Junction

D. M. Fambrough and R. E. Pagano

A few years ago it was discovered that
some membrane proteins are not fixed in
their positions but can move about in the
plane of the membrane, partly by means
of Brownian motion. Lipid molecules
have likewise been found to move. In-
deed, the biological membranes are now
considered to resemble liquid films more
than rigid two-dimensional lattices. How-
ever, certain functional molecules in the
membrane remain confined to a limited
area. The acetylcholine receptors at the
neuromuscular junction seem to be among
the molecules with limited '^translational
mobility." During the past year we have
begun to investigate the restriction placed
by the muscle upon the mobility of ACh
receptors at the neuromuscular junction,
and the underlying mechanism of this
restriction. We can report that the re-
striction upon receptor movement is ex-
treme. ACh receptors do not change their
positions significantly in the plasma
membrane over an 18-hour period at
37 °C in the living tissue.

The design of our experiments is simi-
lar to that first used by Poo and Cone
(Xature 247, 438, 1974) in studies of the
mobility of rhodopsin in visual cell mem-
branes and now being exploited in several
laboratories, including that of Webb at
Cornell. We prepared a fluorescent deriv-
ative of a-bungarotoxin (the tetramethyl-
rhodamine derivative in this case) and
exposed a rat diaphragm muscle in organ
culture to the toxin for 2 hours. Then the
unbound toxin was rinsed away, and the
muscle was mounted for observation with

epi-illumination in the fluorescence mi-
croscope. For illumination we used the
568.1-nm line of a krypton-argon laser;
the beam passed through the objective
lens, a 40X water immersion lens, and
down onto the specimen. With this illumi-
nation, diminished 1000-fold by neutral
density filters, we could locate the neuro-
muscular junctions, which fluoresced
brightly because of the fluorescent «-
bungarotoxin bound to the acetylcholine
receptors there. Neuromuscular junctions
in the surface layer of muscle fibers were
chosen for study and photographed. A
pinhole was placed in the path of the
laser beam so that only an area about 6
^m in diameter was illuminated. The
neutral density filters were removed from
the light path for 3 minutes and the high
intensity laser light allowed to photo-
bleach the illuminated area. Then the
neutral density filters were reinserted in
the light path, the pinhole removed, and
the junction rephotographed. The muscle
was returned to the 37 °C incubator and
maintained in culture for 18 hours. Finally
the muscle was positioned on the micro-
scope again, and the partially bleached
neuromuscular junctions located and
photographed (Fig. 13). If ACh receptors
were free to move about in the junctional
membrane, then during the 18-hour cul-
ture period after partial bleaching the
receptors should have become reposi-
tioned and the fluorescence deficit created
by the bleaching microbeam should have
become distributed over the fluorescent
area. This did not occur. The boundary
of the bleached area remained as sharp
as it had been initially, indicating lack
of movement of receptors across this
boundary.

It is certain that the forces restricting
receptor mobility do not include covalent
bonds to receptor molecules, because the
receptors can be solubilizcd from the
junction })y non-ionic detergents. Some
weaker bonding forces must be involved.
There are agents that specifically affect
certain types of weak molecular inter-
actions, and there are agents that disrupt

DEPARTMENT OF EMBRYOLOGY

29

Fig. 13. Fluorescence micrograph showing the postsynaptic membrane of a rat diaphragm
neuromuscular junction after incubation of the muscle with tetra-methyl-rhodamine-a-
bungarotoxin. Fluorescence is from labeled a-bungarotoxin boimd to ACh receptors. Small
black and white squares ring a dark, disc-shaped area in which the fluorescence has been
bleached by an intense laser beam. The bleaching was done 18 hr before this photograph was
taken. In the interim the diaphragm was cultured in Trowell medium at 37 °C in a humidified
incubator with 95% 02/5% CO2. Magnification bar represents 10 /u,m.

particular cellular structures. We are
using such agents to probe the nature of
the forces restricting receptor movement.

Studies on Chick Muscle
Acetylcholinesterase

R.Rotundo

Acetylcholinesterase is the enzyme that
catalyzses the hydrolysis of acetylcholine
released at the neuromuscular junction
and at other cholinergic synapses in the
nervous system. Several forms of this
enzyme have been described for each
species examined (including the house
fly, electric eel, rat, mouse, chicken, and
ox) . These forms can be distinguished by
their mobility during polyacrylamide gel
electrophoresis and velocity sedimenta-
tion in sucrose gradients. Studies on puri-
fied acetylcholinesterase isolated from
the electric organs of the eel (Electro-
phorus) by Rieger and co-workers, indi-
cate that the various molecular-weight
forms of the enzyme are multimers with
a common subunit structure. However,

several forms of this glycolipoprotein
have also been shown to be attached to a
three-stranded collagen-like ^'tail."

In both the rat and the chick, three
molecular forms can be distinguished ac-
cording to their sedimentation coefficients.
These will be referred to as the high,
intermediate, and low molecular weight
forms of the enzyme. The high molecular
weight form appears to be localized pri-
marily at the neuromuscular junction,
whereas the intermediate and low mo-
lecular weight forms appear in all neural
tissues (Hall, J. Neurobiology 4, 343,
1973; Vignay et al, FEES Letters 69,
277, 1976). Some evidence suggests that
the intermediate form is localized in or
on the external cell membrane. Another
aspect of acetylcholinesterase in muscle
is its apparent trans-synaptic regulation
by the incoming nerve. The high molecu-
lar weight form disappears following
denervation, which makes this enzyme
useful as a tool to study mechanisms of
neuron-target cell interactions.

What we now have is an enzyme mole-
cule that is easily quantified by sensitive

30

CARNEGIE INSTITUTION

radiometric assays and is associated, at
least in part, with the muscle cell mem-
brane. It exist:? in several easily distin-
guishable forms which could be corre-
lated with subcellular localization and
which appear to be under neuronal regu-
lation. We are investigating the proper-
ties and behavior of acetylcholinesterase,
both in vitro and in vivo, to gain a better
understanding of neuronal regulation of
protein synthesis and gene expression in
target cells. Furthermore, the study of
the biosynthesis, release, and degradation
of acetylcholinesterase will yield infor-
mation about the metabolism of a second
membrane protein, and that information
can be compared with data on the me-
tabolism of the acetylcholine receptor.

Properties of Chick Muscle
Acetylcholinesterase

Acetylcholinesterase extracts were pre-
pared by homogenizing and then cen-

trifuging 19-day-old chick embryo leg
muscle in a phosphate buffer containing
Triton X-100, NaCl 0.15 M and EDTA.
A fraction of the supernatant was lay-
ered on a 5-20% sucrose gradient and
centrifuged for 17 hours at 38,000 rpm,
using an SW41 rotor. The results of this
procedure are illustrated in Fig. 14. The
fractions labeled I, II, and III comprise
the heavy, intermediate, and light forms
of the enzyme, respectively. The frac-
tions labeled IV do not appear to be
acetylcholinesterase. That each of these
peaks is a distinct form of the enzyme
rather than an artifact of the preparatory
procedure can be demonstrated by dia-
lyzing each of the pools and running
them on separate 5-20% sucrose gradi-
ents. Figure 15 illustrates the results of
such an experiment using the molecular
forms of the enzyme which had been
frozen for 25 days following initial sepa-
ration by velocity sedimentation. The
forms of the esterase obtained by this

20 30 40

Fraction number

50

60

Fig. 14. Molecular forms of chick muscle acetylcholinesterase. An extract of 19-day-old
rhick embrv'o leg muscle was prepared as described in text. A 500-^1 aliquot was layered on
a 5-207^ sucro.se gradient and centrifuged for 17 hr at 38,000 rpm in an SW41 rotor. Fraction
I is at the bottom of the gradient.

DEPARTMENT OF EMBRYOLOGY

31

<

LjJ

<

20 30 10

Fraction number

30

Fig. 15. Molecular forms of chick muscle after
AChE isolation. The peak fractions illustrated
in Fig. 14 were pooled and stored frozen for 1
month. The fractions were then dialyzed against
homogenization buffer with Triton X-100 and
200-/u,l aliquots centrifuged on 5-20% sucrose
gradients. During storage and preparation all
four forms retain their sedimentation charac-
teristics.

method have been partially characterized
and the results summarized in Table 1.
Although all four forms exhibit the same
apparent Km for acetylcholine, only the
first three appear to be true cholinesterase
or acetylcholinesterase, which is distin-
guished from butyrylcholinesterase by

its ability to be inhibited by BW284c51,
a specific AChE inhibitor, at 10— "^ M.
Additional studies have indicated that
form I, the high molecular weight form,
is preferentially solubilized from tissue
by high salt concentrations (1 M NaClj.
Forms II and III, the intermediate and
low molecular weight forms, are prefer-
entially extracted with solutions contain-
ing Triton X-100, although large amounts
can be obtained by homogenizing in buf-
fer alone. Studies using concanavalin A,
a plant agglutinin binding specifically to
a-mannoside residues, indicate that both
the intermediate and low molecular
weight forms are glycoproteins. Prelimi-
nary evidence suggests that the high
molecular weight form is also a glyco-
protein, but a conclusive study is lacking
because of the instability of this form
under the conditions used to precipitate
AChE with concanavalin A.

In cultures of chick embryo muscle
cells, the high molecular weight form of
AChE is lacking, for there are no neurons
(and thus no neuromuscular connections)
in this preparation (Fig. 16). During
early development in culture only the in-
termediate and low molecular weight
forms are present; but with increasing
time in culture, small amounts of form
IV become apparent. Since it is important
for us to understand precisely the rela-
tionships between the various forms of
the enzyme in order to study their bio-
synthesis and degradation, we are ex-
ploring several methods of affinity chro-
matography that will enable us to obtain

TABLE 1. Properties of Acetylcholinesterase Molecular Forms

AChE Form

Km (mM)

% Inhibition
by 10-' M BW284c51

Con-A Precipitable

I

n
III

IV

0.61
0.60
0.52
0.64

53
51
52
13

N/A

Yes

Yea

N/A

The AChE forms are numbered according to their order of appearance in Fig. 1 : form I
exhibits the highest sedimentation coefficient. The partially purified forms of AChE illustrated
in Fig. 2 were used for the kinetic studies, and the Km values were calculated by regression
analysis of the Lineweaver-Burke plots.

32

CARNEGIE INSTITUTION

rO

1

14

-

g

X

12

-

E

A

o

10

-

M

>%

/ 1

♦-

B

-

i 1

>

[ 1

.,_

1 T

u

e;

-

/ 1

<

1 \

UJ

/)

1 \

x:

• 1

U

/ T

<

»

_ / \

■^

I [^■^•••i*^ I

20 30

Fraction number

40

Fig. 16. Velocity sedimentation of AChE
from chick embryo muscle cultures. Eleven-day
embryo leg muscle cells were cultured for 3
days and then washed and extracted as de-
scribed for whole muscle. Aliquots of 200 lA
were centrifuged on 5-209f sucrose gradients as
in Fig. 14. The molecular forms of AChE illus-
trated here have sedimentation coefficients
identical to the forms labeled II and III in
Fig. 15. Note the absence of forms I and IV in
this preparation.

the enzyme in pure form for further
characterization.

Cell Surface versus Internal AChE

We are attempting to elucidate the
subcellular distribution of acetylcho-
linesterase molecules in cultured chick
embryo muscle cells and to examine both

the relationships between the various
molecular forms of the enzyme and their
metabolic fates. A simple assay system
was devised to test for the presence of
AChE on the external plasma membranes
of cultured muscle cells and to determine
the ratio of extracellular to intracellular
AChE. i^C-acetylcholine in a buffered
salt solution is added to muscle cultures
that have been washed to remove all
traces of medium esterases. The distribu-
tion of ^*C-acetylcholine in the assay sys-
tem (Table 2) indicates that the labeled
substrate does not penetrate the cells.
Thus any hydrolysis of ^'^C-acetylcholine
to ^^C-acetate must occur by AChE out-
side the cells. The distribution of ^^C-
acetate shows that most of the labeled
product also remains in the medium. By
removing and counting the ^^C-acetate
produced by the hydrolysis of ^*C-acetyl-
choline in the medium over a given time
interval, a measure of AChE activity is
obtained. After an initial incubation
period to measure AChE on the cell sur-
face, enough Triton X-100 can be added
to make a 1% -detergent medium. This
procedure solubilizes the cell membranes
and renders the labeled substrate accessi-
ble to all the AChE in the cell. After
Triton X-100 treatment, a second meas-
urement of AChE activity gives the total
enzyme content of the cells. Using this
double incubation procedure, we have
been able to measure AChE activity both
on the cell surface and within the muscle

TABLE 2. Distribution of "C-Acetylcholine and "C-Acetate on Cultured Muscle Cells

"C-ACh

^*C-Acetate

CPM

% Total

CPM

% Total

Total CPM

Assay medium
Cultured cells

Total

663360
30

663390

100

99150
2800

101950

97.3
2.7

762510
2830

765340

Total CPM recovered

765340

= 99.7%

Total CPM added 767760

Cultures were as.sayed for surface AChE as described in the text. After incubation the
medium was removed and the cells extracted in Triton X-100 buffer solution. The "C-acetyl-
choline was separated from the ^*C-acetate by sodium tetraphenylboron (Kalignost) in 3-
heptanone. Each value is the mean of three determinations.

DEPARTMENT OF EMBRYOLOGY

33

a>

O
>

>
o

<

UJ

JZ

o

<

200

-

.

160

-

/

-

i ^^

- "^

120

-

I ^^^ D --

J^-^

80

-

y

40

Cy^

/^ 1

i+TX-lOO or removal

of medium
1 1 1 1

1

20

40 60 80 100

Time ( minutes )

120

Fig. 17. Cell surface AChE assay. All values have been normalized to "C-ACh CPM hydro-
lyzed per hour. (O-O) Cultures were incubated for 1 hr at 37 °C, at which time the medium
was made 1% with respect to Triton X-100 and incubated for another 15 minutes. (A-- A)
Cultures were incubated for 1 hr at 37 °C and then removed from the medium, which was
then incubated for an additional 45 min in the absence of cells to demonstrate the presence

of AChE released into solution during the assay. (D D) Cultures were incubated at 5°C

for 1 hr and then removed from the medium, as described for the preceding experiment. Post-
incubation of the medium at 5°C shows that no AChE was released into the medium under
these conditions. Each point is the mean of three cultures.

cells. An example of this kind of experi-
ment is illustrated in Fig. 17. Such
studies indicate that approximately 20%
of the AChE is localized on the external
cell membrane. As will be discussed in a
subsequent section, however, AChE is
also released into the medium when mus-
cle cells are incubated at 37°C. To cir-
cumvent this difficulty, we assayed for
cell-surface AChE activity by incubat-
ing at 5°C and then removing the medium
from contact with the cells and con-
tinuing the incubation. Figure 17 illus-
trates one such experiment, which shows
that at this temperature there is no
release of AChE from the cells. When
muscle cells are incubated for 1 hour at
20°C, they release approximately 10%
as much AChE as is found on the mem-
brane.

Table 3 illustrates the effects of tem-
perature and specific inhibitors on the
rate of acetylcholine hydrolysis by cell-

surface AChE. At 37°C both cell-surface
AChE and newly released AChE contrib-
ute to the measured rate of acetylcholine
hydrolysis, whereas at 20°C the contribu-
tion of the new AChE is 90% that of the
cell-surface enzyme. The lack of inhibi-

TABLE 3. Effects of Temperature and Specific
Inhibitors on Cell Surface AChE Activitv

Treatment Temp. °C

AChE activitv

(% 37° Control)

None

37

100

None

20

50

10-* M IsoOMPA

37

98

10"" M BW284c51

37

20

Chick embryo muscle cultures were incubated
for 2 hours as described in text. BW284c51 at
lO"'' M is a specific inhibitor for acetylcholin-
esterase; IsoOMPA at 10"* M is an inhibitor
specific for butyrylcholinesterase. At 37 °C the
AChE activity measured includes both the
molecular form on the outer cell membrane and
those released into the medium.

34

CARNEGIE INSTITUTION

tion by 10-^ M IsoOMPA. a butyryl-
cholinest erase inhibitor, and the 75%
inhibition by 10"^' .1/ BW284c51. an
acetylcholinesterase inhibitor, at 37°C
indicates that both the cell-surface and
the newly released forms of the enzyme
are true acetylcholinesterases. We are
using: this assay system along with sev-
eral specific reversible and irreversible
inhibitors of AChE to study the kinetics
oi turnover of the intracellular and ex-
tracellular enzyme pools and their rela-
tion to the release of AChE into the
medium. Preliminary experiments sug-
gest that the rate of turnover of cell-
surface AChE molecules and the release
oi AChE into the medium are separate
but metabolically coupled processes.

Release of AChE by Chick Muscle
Cells in Culture

Besides being found both inside the
muscle cells and on the external mem-
brane, substantial quantities of acetyl-
cholinesterase are also released into the
culture medium (Wilson et al., Develop.
Biol. 33, 285. 1973). To study the re-
lease of AChE molecules, however, we
needed a culture medium devoid of
esterase activity. This was accomplished
by treating our regular Eagle's medium
containing 2% chick embryo extract and
10% horse serum with, diisopropylfluoro-
phosphate (DFP) , an irreversible esterase
inhibitor, at a concentration of 2 X 10"'''
M. The excess DFP in solution then
becomes hydrolyzed over the next several
days, and the medium thus treated does
not exhibit cholinesterase activity for at
least two weeks.

To determine how much AChE is re-
leased by cultured muscle cells, the
cultures are first washed in Hank's
buffered salt solution to remove untreated
medium and AChE previously released
into that medium. The cultures arc then
covered with 1-2 ml of DFP-treatcd
medium and returned to an incubator.
At the designated time intervals, 10-//1
aliquots of medium arc sampled from

each culture and stored at 5°C. After
the last sample is obtained the tubes are
assayed for AChE activity. Pharmaco-
logical agents, when they are used, are
added directly to the medium or diluted
and added in a volume of 10 ^1/ml
medium. All manipulations are performed
at 37°C.

Figure 18 illustrates the effect of the
inhibition of protein synthesis on the
rate of AChE release into the medium.
Both puromycin (10 ^g/ml) and cyclo-
heximide (100 /^g/ml) inhibit the release
within approximately 3 hours. This indi-
cates that the release of previously syn-
thesized AChE requires the synthesis of
specific proteins, or that the AChE re-
leased into the medium is newly synthe-
sized and the cells contain a pool of
AChE molecules equivalent to that re-
leased over 2-3 hours. Evidence obtained
from the density shifting technique, to
be described subsequently, favors the
second hypothesis. The release of AChE
is also temperature-dependent and re-

UJ

<

500

-

400

-

O/^

-^

300

-

yo

200

-

q/ ^--a--^~-
/,r-'^'--D---tr--c>--

100
<

aV

1 1 I 1

•

-*

1

2 4 6

Time ( hours

10

Fig. 18. Effects of metabolic inhibitors on the
relea.se of AChE into the medium. All values
have been normalized to the amount of AChE
released by control cultures during the first
hour. (O-O) untreated controls; (A- -A) 10
/xg/ml puromycin; (n--n) 100 /^g/ml cyclo-
heximide; (A-A) 10"^' M carbonyl(;yanide-m-
chloroj)henylhydrazone; (•-•) lO"""' M dinitro-
phenol. Each ])oint is the mean of three cul-
tures.

DEPARTMENT OF EMBRYOLOGY

35

quires energy in the form of ATP, as
shown in Fig. 18. Here the release of
enzyme is completely blocked after a 1-
hour exposure to 10~"^ M dinitrophenol
or 10~^ M carbonylcyanide-m-chloro-
phenylhydrazone. These effects of ATP
and protein-synthesis inhibitors on the
release of AChE as well as the time-
course of effect correspond well to those
observed for the synthesis and incorpora-
tion into membranes of the acetylcholine
receptor and suggest that a common
mechanism might be involved.

To test this hypothesis further we
treated chick embryo muscle cultures
with 10-4 j^ J3FP for 15 minutes. This
procedure resulted in essentially complete
and irreversible inhibition of both intra-
cellular and extracellular AChE activity.
The cultures were then returned to the
incubator and the rate of reappearance
of AChE activity in the medium moni-
tored as previously described. Fig. 19
illustrates the results of such an experi-
ment. The time from AChE biosynthesis
to release is approximately 2% hours,
which is close to the time for ACh re-
ceptor incorporation into the muscle cell
membrane. The nature of the release
mechanism for AChE and its relation to

acetylcholine receptor biosynthesis are
current topics of investigation.

The enzyme released into th(; medium
is inhibited by lO"-* M BW284c51 but
not by 10-^ M IsoOMPA, a specific
inhibitor of butyrylcholinesterase. This
leads us to a preliminary characteriza-
tion of the enzyme as a true cholines-
tcrase. Like the AChE forms found in
the cells, the released AChE is pre-
cipitable by concanavalin A, and this
indicates that it is a glycoprotein as well.
However, during velocity sedimentation
experiments in 5-20% sucrose gradients,
the released form of the enzyme migrates
as a broad band with an S value midway
between those of the two cellular forms.
These data suggest that it exists as a
distinct form, possibly a modification of
one of the other two. These differences
between the cellular and released forms
are illustrated in Fig. 20.

I

o

£

Q.
O

>%
">

U

<

LlI

.c
O
<

O

£
o.
o

<

LiJ
<

E
-a

2 4 6 8

Hours post -treatment

Fig. 19. Effects of DFP on the release of
AChE in chick embryo muscle culture. (O-O)
untreated controls; (A- -A) cultures treated
with 10-* M DFP. Each point is the mean of
three cultures.

Fraction number

Fig. 20. Velocity sedimentation of medium
and cell AChE from chick embryo muscle cul-
tures. (•-•) cell AChE; (0--0) medium
AChE. In this experiment 300-/ttl aliquots of cell
extract or medium were layered on 5-209<r
sucrose gradients and centrifuged in an SW50.1
rotor at 48,000 rpm for 13 hr. The result of two
separate gradients are plotted on the same
graph.

36

CARNEGIE INSTITUTION

ro
I

O

E

o.

o

>

<

LU
SI
O
<

30

A

25

M • \

20

-

Ui

15

-

i 1 ' 1^

10

-

5

»«f

10 15 20

Fraction number

25

Fig. 21. Isopychnic centrifugation of medium AChE in CsCl. ( ) medium from cultures

exposed to heavy amino acids for 48 hr; ( ) medium AChE after 7-hr exposure of

cells to hght amino acids; ( ) medium AChE after 19-hr exposure of cells to light

amino acids.

Density Shifting of Acetylcholinesterase

We have employed the technique of
density shifting described by Devreotes
and Fambrough (Year Book 75) to
demonstrate the synthesis of new AChE
molecules in cultured chick muscle cells
and the release of newly synthesized
AChE into the medium. In one such
experiment chick muscle cultures that
had been incubated with medium con-
taining D, ^-"^C, ^•''N amino acids for 48
hours were switched to medium contain-
ing "light" amino acids for 7 or 19 hours.
Controls remained in dense medium.
One-milliliter portions of medium were
mixed with 2.37 g CsCl in 4 ml 100 mM
PO4 buffer j)Y\ 7.0 and centrifuged for
89 hours at 40,000 rpm in an SW65

rotor. The results of this experiment are
shown in Fig. 21. Both light and dense
AChE molecules are found in the dense
medium because of the presence of light
esterase derived from the horse serum
and embryo extract. The dense AChE is
that which has been synthesized by the
cells during their exposure to the heavy
amino acid-containing medium. After
the transfer to light amino acid-contain-
ing medium, all of the AChE released
is of the light form, indicating that the
newly released AChE molecules have
been recently synthesized.

Figure 22 shows a similar type of ex-
periment designed to investigate the turn-
over of AChE molecules in cultured
muscle cells. Chick muscle cells were
grown in medium containing the normal

DEPARTMENT OF EMBRYOLOGY

37

I
O

e

Q.
U

>

<

UJ
jC
U
<

10 20 30

Fraction number

Fig. 22. Isopychnic centrifugation of cell ex-
tract AChE in CsCl. ( ) light amino acid

controls; ( ) cell AChE after 17-hr ex-
posure of cultures to heavy amino acids ; ( )

cell AChE after 48-hr exposure of cultures to
heavy amino acids.

light amino acids and then shifted to
heavy amino acids. After the appropriate
incubation time, the plates were washed
to remove medium AChE, and the enzyme
associated with the cells was extracted
with 100 mM phosphate buffer pH 7.0
with Triton X-100, 0.15 M NaCl and
0.25 mM EDTA.

The resulting extracts were then sub-
jected to isopycnic centrifugation in
CsCl, as described above, for 96 hours.
In contrast to the density shift of AChE
in the medium, which was essentially
complete following the change to light
medium, the enzyme associated with the
cells contained both light and dense
AChE molecules even after 48 hours in
dense medium. Taken together, these
studies suggest the presence of at least
two subcellular compartments of AChE:
The first is a pool with rapid turnover
of newly synthesized AChE molecules
which are released into the medium; the
second is a compartment whose molecular
contents are more slowly turning over
and do not appear to leave the cell.

These experiments illustrate the use-
fulness of the density-shift technique in

studying the distribution of newly syn-
thesized AChE molecules in muscle cell
cultures. We can now use this technique
to trace the metabolic fate of the several
forms of acetylcholinesterase found in
muscle cell cultures, and to answer ques-
tions about the possible interconversion
of molecular forms of the enzyme.

Acetylcholine Receptors and

a-BuNGAROTOXIN BINDING SiTES

ON Neurons
S. Carbonetto and K. Midler

a-Bungarotoxin binds specifically to
the acetylcholine (ACh) receptor in
skeletal muscle and has been used very
successfully to study the metabolism of
this membrane protein. The specificity
of a-bungarotoxin for the ACh receptor
has been established by: (1) the localiza-
tion of a-bungarotoxin binding to cells
and regions of cells that respond physio-
logically to ACh, (2) the saturable bind-
ing of a-bungarotoxin to muscle and its
competition by agonists and antagonists
of ACh, and (3) inhibition of the physio-
logical response of muscle to ACh by
a-bungarotoxin.

a-Bungarotoxin binds specifically to
primary cultures of chick sympathetic
neurons as well as the ACh receptor in
muscle. In Year Book 75 we reported
that a study using light autoradiography
had shown that a- [^^^I] bungarotoxin
binding is localized to sympathetic neu-
rons in culture and that this binding is
completely blocked by curare (10~^ M) .
Since then we have confirmed that the
localized application of ACh onto neurons
by microionophoresis results in a de-
polarization of the membrane potential
measured with intracellular microelec-
trodes (Fig. 23). This ACh response is
regarded as unequivocal evidence for the
presence of an ACh receptor in the
plasma membranes of these cells. The
ACh response is quickly blocked by
curare (10—* M) but can still be evoked
in the presence of a-bungarotoxin (1

38

CARNEGIE INSTITUTION

Control

aBuTX

Control

ocBuTX

B

J

dTC

y

Fig. 23. Effect of curare and a-bungarotoxin on the ACh response in neurons. (A) Responses
of two neurons to short pulses of iontophoretically apphed ACh in the absence (control) and
presence of 1 fj.g/m\ of a-bungarotoxin (aBuTX). Upper trace: 200 nA, 10 msec. Lower trace:
10 mV, 10 msec. (B) Response of two neurons to release of the ACh holding current in the
ab.sence (control) and presence of 1 /j.g/m\ of a-bungarotoxin (aBuTX). Upper trace: onset
and off.set of release of the holding current. Lower trace: 10 mV, 5 sec. (C) Response of the
same neuron to short pulses of ACh applied before the addition of 10~* M curare and 35 min
afterwards (dTC) . Upper trace 200 nA, 2 msec. Lower trace : 10 mV, 2 msec.

/zg/ml). Furthermore, preincubation of
neuron.s with a-bungarotoxin does not
affect curare's block of the ACh response.
a-Bungarotoxin also binds specifically
to chick sympathetic ganglia, so the effect
of a-bungarotoxin was tested on an in
vitro preparation of the lumbrosacral
sympathotic chain. Stimulation of the
interganglionic connective of the chain

produced a response, recorded extracellu-
larly from the ganglionic root, that was
blocked reversibly by low concentrations
of curare flO~^ M) and could not be
evoked by antidromic stimulation (Fig.
24). These two results suggest that the
extracellularly recorded '^ganglionic po-
tential" was mediated by cholinergic
synapses, yet a-bungarotoxin in concen-

DEPARTMENT OF EMBRYOLOGY

39

die

\f0^ i\j^^*'m>«km0t*t

a BuTX

Fig. 24. Effect of curare and a-bungarotoxin on ganglionic response of chick sympathetic
ganglia. (A) Ganglionic response before and 5 min after application of 10"* M curare. (B)
Ganglionic response before and 1 hr after application of 10 Mg/ml of a-bungarotoxin. 40 ixV,
40 msec.

trations up to 60 /xg/ml had no effect on
the amplitude of the potential. Thus a-
bungarotoxin satisfies two important cri-
teria that would establish it as a ligand
for the neuronal ACh receptor: It binds
to cells known to have ACh receptors,
and it is competed by antagonists of
ACh. Nevertheless, the failure of a-
bungarotoxin to block the physiological
response to ACh makes us unable to
certify it as a ligand for the neuronal
ACh receptor.

There are several possibilities that
could explain the inability of a-bungaro-
toxin to block the ACh response and still
have it bind to the neuronal ACh recep-
tor. There may be multiple antagonist
binding sites with a-bungarotoxin bind-
ing poorly to its site so that it is incapable
of either inducing an allosteric change
or sterically hindering ACh and curare
from gaining access to their sites, a-
Bungarotoxin binding to neurons is re-
versible and dissociates with a half-time
of 2.5 hours at 23°C, whereas a-bungaro-
toxin binding to muscle is almost irre-

versible. There may be a small popula-
tion of ACh receptors that are blocked by
a-bungarotoxin, but their effect on the
ACh response would be too small to be
resolved by the physiological methods
used.

The molecular weight of the ACh re-
ceptor from skeletal muscle is approxi-
mately 250,000 daltons, and the receptor
sediments at 9.5 S to 11 S in sucrose
gradient centrifugation. As mentioned
previously, the binding of a-bungarotoxin
to sympathetic ganglia is saturable; and
a detergent extract of membranes from
ganglia, when assayed by sucrose gradient
centrifugation, has a distinct a-bungaro-
toxin binding peak that sediments very
similarly to the muscle ACh receptor and
is completely competed by curare (Fig.
25) . This suggests again that the a-bunga-
rotoxin binding molecule in neurons is the
ACh receptor and, furthermore, that
there is a detergent-soluble a-bungaro-
toxin "receptor" that can be used to
study membrane turnover in neurons.

40

CARNEGIE INSTITUTION

CL
O

3
GO

600

-

r\

500

-

/ \ \

400

^ ' /'"^ \

300

-

/ 1 / • ' \

200

I My •■ \ \

100

-

^ k / \ \

•-

-A.

iJ:

10 20

FRACTION NUMBER

30

Fie. 25. Velocity sedimentation of neuronal a-biingarotoxin receptors. Six lumbosacral
ganglionic chains were dissected from 19-day chick embryos and homogenized to 10 mM
Tris, 1 m.V PMSF-EDTA (5 ml). The homogenate was centrifuged at 27,000sf for 60 min and
the resulting membrane pellet extracted in 700 ti\ 1% Triton X-100 1 niM PMSF-EDTA.
This was centrifuged again at 27,0009^ for 60 min, and the supernate was divided into 200-/ul
aliquots. Each aliquot was preincubated with 10"*M curare or an equal volume of water for
20 min at 23^C and then incubated with 0.01 /xg/ml of a-r^''T]bungarotoxin for 45 min at
37 "C. Free a-bungarotoxin was removed by chromatography on a Biogel P-60 column (5
ml), and the excluded peak (approximately 250 lA) was layered in a thin band over a 5-20%
linear sucrose gradient (5 ml) containing 1% Triton X-lOO, 1 niM PMSF-EDTA. 15 ^1 of
muscle ACh receptors labeled with a-[^"T]bungarotoxin was also layered on each gradient
as a marker. Centrifugation was caiTied out in a Beckman SW50.1 rotor at 48,000 rpm for
6 hr at 8°C. Gangha (•-•); ganglia +dTC (• •) ; marker (A--A).

Binding of a-bungarotoxin to sympa-
thetic neurons, however, is rapidly re-
ver.«ible. and this presents an o!)stacle
for turnover studies that would rely on
long periods of ultracentrifugation to
assay the receptor. Fortunately, we have
found that glutaraldehyde in concentra-
tions as low as 0.01% can he used to
covalently cross-link a-})ungarotoxin to
its receptor and decrease its dissoci-
ation from 50% in 3.5 liours to less than
10% after 24 hours at 23°C. Cross-linked
bungarotoxin-receptor complexes sedi-
ment with an »S value similar to that of

muscle ACh receptors (Fig. 26). The free
neuronal a-bungarotoxin-binding mole-
cules also sediment as approximately IOaS
molecules, and their position in sucrose
gradients can be determined by a-bunga-
rotoxin binding.

We are embarking upon experiments
to study the metabolism of the a-bunga-
rotoxin binding molecule in neurons by
means of the density-shift technique.
These experiments may make an im-
))ortant contribution to our knowledge of
the tui'nov(>r of neuronal membranes and
neuronal growth.

DEPARTMENT OF EMBRYOLOGY

41

200 -

CL
O

3

GO
a

100 -

20
FRACTION

30
NUMBER

Fig. 26. Velocity sedimentation of cross-linked a-bungarotoxin receptors. Ten lumbo-
sacral ganglionic chains were dissected from 13-day embryos and pinned out in a Sylgard
(Dow) lined plastic culture dish. These were incubated with 0.03 /Wg/ml of ^T-labeled a-
bungarotoxin for 60 min at 37 °C and washed thoroughly. The ganglia were then homogenized
in 450 ^A of 1% Triton X-100, 1 mM PMSF-EDTA in a ground glass homogenizer and
spun for 45 min at 27,000gr. 200X aliquots of the supernate were treated with V/c gluteraldehyde
for 20 min at 23 °C and were layered in a thin band over a 5-20% linear .sucrose gradient (5
ml) containing 1% Triton X-100, 1 mM PMSF-EDTA. 40 /^l of muscle ACh receptors labeled
with ^^^I-a-bungarotoxin was also layered on each gradient as a marker. Centrifugation was
carried out in a Beckman SW50.1 rotor at 48,000 rpm for 5 hr at 5°C. Ganglia (•-•) ;
marker (A- -A).

ORGANIZATION AND INTERACTIONS OF MEMBRANE
LIPIDS IN MAMMALIAN CELLS

R. E. Pagano, A. Sandra, L. Huang, K. Ozato, and M. Takeichi
with the assistance of E. Asch, W. Duncan, B. Smith, and J. Wiser

Studies in this laboratory continue to
focus on the dynamic role which mem-
brane lipids and cholesterol play in cell
membrane structure and function. The
generally accepted concept of the cell
plasma membrane is the fluid mosaic
model of membranes in which the matrix
of the membrane is a lipid bilayer and
in which proteins are embedded to vary-
ing degrees and are free to undergo
lateral diffusion. It is thought that under

certain physiological conditions portions
or domains of this lipid matrix can
change their physical state and thus alter
the functional properties of the mem-
brane. There is also a growing body of
evidence, derived from studies of bac-
terial and erythrocyte membrane sys-
tems, that the various phospholipid spe-
cies in membranes may be asymmetrically
dispersed across the transverse dimension
of the bilayer. In this report we examine

42 CARNEGIE INSTITUTION

the transbilayer distribution of the major preparations of plasma membranes which

cellular phospholipids in the }")lasma are pure, sealed, and of uniform sided-

membrane of cultured mouse L^I cells ness. In our studies we have used isolated

and extend the notion of membrane lijnd latex phagosomes. These phagosomes

asymmetry to this complex eukaryotic have a phospholipid composition similar

cell type. to that of isolated plasma membranes,

In other studies we are examining the and from the topography of the phago-

interactions of artificial lij^d vesicles cytotic process, we assume that an inside-

with mammalian cells. Our approach to out membrane sidedness is preserved,

understanding the physical and biochem- L]\I cells were grown in tissue culture

ical properties of natural membranes has in medium supplemented with '"^H-choline,

been to study synthetic lipid vesicles of "^H-palmitate, -"^H-acetate, or ^''^P-ortho-

known composition and structure. In i)hosphate. The cells were washed, sus-

principle, it should be i^ossible to isolate i)ended in medium, and incubated from

electrostatic, structural, and dynamic 15 to 60 minutes with 3000 polystyrene

contriinitions to the interaction process latex beads (0.5 /x-2.02 /x diam.) per cell,

by controlled and systematic variation The ingested beads were harvested by

of the chemical and physical properties homogenization of the cells and isolated

of the synthetic membrane surface. This by centrifugation on buoyant density

year we report on the role of the cell- sucrose gradients. Figure 27 shows the

surface proteins and vesicle lipid com- distribution of radioactivity using ^H-

position in the adhesion of lipid mem- choline labeled cells. It is seen that the

branes to the cell surface. ^H-cpm comigrates with the position of

Details of our studies on each of these the isolated latex beads. No peak of

topics are presented in separate sections radioactivity was observed at this posi-

below. tion in labeled cells homogenized in the

absence of beads. More than 99% of the

^ _ _ radioactivity was chloroform-methanol

STUDIES OF Traxsbilayer DISTRIBUTION ^xtractablc and more than 90% of the

OF Phospholipids in ^Iammalian Cells radioactivity was in PC and SM. Iso-

A. Sandra and R.E.Pagano ^^ted latex phagosomes were found to

be enriched for the plasma membrane
We examined the distribution of the marker 5' nucleotidase and devoid of
major phospholipid classes (PC, phos- LDH activity, lending further support
phatidyl choline; PE, phosphatidyl etha- for the plasma membrane origin of the
nolamine; PS, phosphatidyl serine; PI, phagosome membranes,
phosphatidyl inositol; and SM, sphingo-
myelin) across the transverse dimension Asymmetry of the Major Phospholipid
of the plasma membrane lipid bilayer of Species (PC andPE)
cultured mouse LM cells. That is, what Phosphatidyl choline phospholipid ex-
traction of each of these species faces change protein (PC-PLEP) from beef
the cytoplasm of the cell versus the liver was isolated by DEAE-52, CM-52,
external bathing medium? Because the and Sephadex G-50 column chromatog-
acyl chain distribution in each of these ^aphy. This protein specifically catalyzes
lipid classes is heterogeneous, we also the 1:1 exchange of PC from donor to
.set out to determine the fatty acid com- acceptor membranes. From studies of
position on each half of the bilayer. ^q^j^I bilayer membranes, it has been

„T ^r I r^ .I- showu that PLEP acts only on the ex-

rlasyna Memtjrane Derivatives , , •, , j. rxi i-i

ternal, or accessible, aspect of the bilayer.

To assess the transbilayer distrilmtion We have used this protein to deter-

of phospholipids, it is necessary to obtain mine the proportion of PC in the accessi-

DEPARTMENT OF EMBRYOLOGY

43

*i**t««i***

■ I ■ ■ -*

BOTTOM

TOP

FRACTION

Fig. 27. Isolation of latex beads from ^H choline-labeled LM cells. (A) Buoyant density
sucrose gradient of bead-containing cell homogenate. Arrows represent position of latex beads;
(B) isolated latex beads rerun under identical conditions; (C) ^H choline-labeled LM cell
homogenate without latex beads run as (A).

ble pool of ^H-choline-labeled latex
phagosomes derived from uniformly la-
beled LM cells. The phagosomes were
incubated with and without PLEP in
the presence of excess DOL (dioleoyl
lecithin) unilamellar vesicles. After the
reaction is terminated, the vesicles and
beads are separated by centrifugation
and their lipids analyzed by thin-layer
chromatography and liquid scintillation
spectrometry. Figure 28 shows the time-
course of the exchange. About half
(^55%) of the PC is readily available
for exchange and is assigned as the cyto-

plasmic component of PC in the LM cell
plasma membrane. The remaining 45%
of PC, which is not available for ex-
change, presumably occupies the inner
portion of the phagosome membrane, i.e.,
the outer portion of the cell plasma
membrane.

To determine the PE asymmetry in
membranes, a nonpenetrating probe that
reacts covalently with primary amino
groups of lipids and proteins, TNBS
(trinitrobenzene sulfonic acid) was used.
Isolated ^^P-labeled latex phagosomes
were reacted with TNBS by incubating

44

CARNEGIE INSTITUTION

6 9 12

TIME (HRS)

18

Fig. 28. Kinetics of ^H PC exchange between
unilammelar dioleoyl-phosphatidyl choline vesi-
cles and ■''H choline-labeled latex phagosomes.
Beads and vesicles were incubated at 37° with
20 U/ml of PC exchange enzj^me. The ex-
changed phospholipids were determined by ex-
traction. TLC separation, and radioactivity
measurement with and without the exchange
protein.

the beads at either 4° or 37°C in Hanks'
balanced salt solution, buffered with
l.S^/c NaHCOa, pU 8.5. The reaction was
stopped by the addition of 0.5 A' HCl or
complete LM medium. Beads were cen-
trifuged and washed with the same solu-
tion. The lipids from the TNBS-labeled
phagosomes were extracted and subjected
to thin-layer chromotography. The Tnp-
PE derivative was readily separable
from authentic PE, and by radioactive
counting procedures the fraction of modi-
fied PE could be readily determined.

The TXBS-labeling experiments were
also performed on intact LM cells, which
were subsequently allowed to ingest latex
beads, -^^p-iabeled LM cells were labeled
with TXBS (5 mM) by suspending the
washed cells in Hanks' BSS containing
1.8% NaHCO, at pH 8.5. The reaction
was stopped by the addition of LM me-
dium. The cells were washed, resuspended

in LM medium, and incubated at 37 °C
with 0.945 /t latex beads for 30 minutes.
The beads were isolated from the cells as
previously described and then analyzed
for PE and Tnp-PE. Thus we were able
to obtain latex phagosomes labeled by
TXBS from either aspect of the plasma
membrane.

TX^BS labeling of isolated latex beads
was found to be temperature dependent.
At 4° about 70% of the PE of the iso-
lated phagosomes was converted to the
Tnp derivative. Labeling at 37 °C in-
creased this value to more than 98%
(Fig. 29). When cells were first labeled
by TX^BS, the ingested latex beads had
only about 24 7r of the PE converted to
Tnp-PE. By raising the temperature of
TX^BS labeling to 37°C, this amount was
increased to 757^ over 120 minutes.

These experiments indicate that PE
is asymmetrically distributed in the
phagosome membrane, with about 70%
occupying the cytoplasmic pool and
about 24% on the outer half of the bilayer.

PE and PC are the major phospholipids
found in the LM cell phagosome. To-
gether they account for about 75% of
the total phospholipid. By knowing the
relative abundance of the phospholipid
species and assuming that total phospho-
lipid is distributed equally on both halves
of the membrane, we deduce the asym-

60
TIME (MIN )

120

Fig. 29. Kinetics of TNBS-labeling of intact
LM cells (•-•) and isolated latex phagosomes
(O-O) as a function of temperature.

DEPARTMENT OF EMBRYOLOGY

45

PC

I

SM

I

PS, PI

I

finiifinni fiiififffiiiiii! tmm "-""

iiiiiiiuiiiiiiu m mmmmm

PE

CYTOPLASMIC
FACE

Fig. 30A. Transbilayer distribution of the phospholipid species of LM cell lafex phagosomes.

metric distribution of the major lipid
species to be as shown in Fig. 30 A. PE
preferentially occupies the inner leaflet
of the bilayer, while the distribution of
PC is more nearly symmetrical. The re-
maining membrane phospholipids, sphin-
gomyelin, phosphatidyl serine, and phos-
phatidyl inositol, preferentially reside on
the outer face of the plasma membrane.

Acyl Chain Asymmetry

The fatty acid portions of the phago-
some phospholipids in each half of the
bilayer were analyzed by first isolating
the exchangeable and nonexchangeable
pools of PE and PC as described previ-
ously. The cellular phospholipids in these
experiments were radiolabeled in the
fatty acid portion by '^H acetate.

The individual phospholipids were hy-
drolyzed, trans-esterified, and subjected
to gas-liquid chromatography. Fractions
of the effluent were then analyzed for
radioactivity. The four major fatty acids
(16:0, 16:1, 18:0, 18:1) which make up
more than 959^: of phagosome acyl chains
were expressed in terms of their relative
abundance. No major differences w^ere
observed in the transbilayer distribution
of acyl chains in PE and PC. The total
acyl chain distribution in SM, PS, and
PE was also determined. In our study of
phospholipid polar headgroup asym-
metry, 22% of the outer face phospho-
lipid and 4% of the cytoplasmic face
material is left over to be assigned to
these species. Because SM is the major
phospholipid to be fitted into the scheme,
we tentatively assign it to the outer face,

CO

<

I
o

_J
>

<

Q

a.

_i
o

I

Q.
CO
O

X

30

20

10

<

O 20

30

16=0 I6:| 18:0 I8:|

B

OUTER
FACE

CYTOPLASMIC
FACE

Fig. SOB. Transbilayer distribution of the
phospholipid acyl chains in LM cell phagosomes.

and PS and PE to the remaining cyto-
plasmic and outer portions. Thus we
are able to construct an acyl chain distri-
bution based on phospholipid asymmetry
and the relative abundance of the acyl
chain distribution within each phospho-
lipid species. These data are summarized
in Fig. 30 B. The cytoplasmic portion of
the plasma membrane is enriched in un-
saturated fatty acids (16:1 and 18:1)
compared to the outer portion. Because
the physical state of phospholipids is
partly determined by the polar head
group and the acyl chain composition,
these results suggest that each half of
the plasma membrane lipid bilayer may
respond differentially under certain phys-
iological conditions. Experiments are in
progress to test such a hypothesis.

46

CARNEGIE INSTITUTION

Adhesion of Phospholipid ^Membranes

TO Chinese Hamster Fibroblasts:

Role of Cell-Surface Proteins

R . E. Pagano and M . Takcichi

In previous reports we have docu-
mented the ability of lipid vesicles 250-
500 A in diameter to fuse and undergo
lipid exchange reactions with the plasma
membranes of Chinese hamster V79
fibroblasts. Here, we show that unilamel-
lar lipid vesicles generated from dimyri-
stoyl lecithin (D^H.) or dipalmitoyl
lecithin (DPL) are taken up at tempera-
tures below their T,.* by V79 cells in
suspension primarily through stable
adsorption or adhesion of vesicles to the
cell surface. It is further demonstrated
that this vesicle-to-cell adhesion requires
the presence of several cell-surface pro-
teins that have a high affinity for ''rigid"
vesicle lipids.

Chinese hamster V79 cells were grown
as monolayer cultures and dissociated
into single cells, using EDTA. Washed
cells were incubated for various periods

* To. gel-liquid crvstallinc phase transition
temperature: T.. ^20.9'C for DML and 36.4°C
for DPL.

of time in a serum- and lipid-free medium
with ^^H DAIL or -^H DPL vesicles.
Typical results are summarized in Table
4 for a 1-hour incubation. The absolute
amounts of DPL uptake at 2°C are
2-2.5 times greater than that at 37°C.
DPL uptake was always greater than
when DjNHj was used. Preincubation of
cells (30 min, 37°C) with 5 mil/ NaNs
and 50 mM deoxyglucose has no effect
on DPL uptake but reduces DML up-
take at 37°C to about 80% of control
values. The presence of 1 mil/ Ca-+ and
1 mM ]\Ig-+ during the vesicle-cell in-
cubation results in a slight (^5%) en-
hancement of vesicle uptake.

The exogenous phospholipid that be-
comes cell-associated upon incubation
with DlNHv or DPL vesicles can be re-
leased by treatment with trypsin. Cells
were incubated for 1 hour with DPL or
DAHj vesicles and subsequently washed
free of excess vesicles. Samples of the
cells were then subjected to a 10-minute
incubation with a 0.01% solution of crys-
talline trypsin at 22 °C. The reaction was
stopped by addition of a tenfold excess
of trypsin inhibitor and by centrifuga-
tion of the cells. Analysis of the cell pellet

TABLE 4. Comparison of DPL and DML Uptake by EDTA-Dissociated (E) and Trypsinized

(T) V79 Fibroblasts in Sus}iension

Lecithin

Uptakcf

/xg/4 X 10"

' cells/hour

With M(

?tabolic

Inhibitors!

Control

(% of Control)

Vesicle-Cell

Cell Type*

Incubation (°C)

DML

DPL

DML

DPL

£

2

18.0

50.0

96

110

E

37

6.5

20.1

80

110

T

2

7.5

19.3

105

109

T

37

7.0

12.1

100

99

* E-cells were prepared from 8ubconfluent monolayers by dissociation with EDTA, T-cells by
combined use of EDTA and try])sin.

t All experiments were farrie*! oul in du))licale or liii)li('alo. Variation was within ±5%.

t Cells were preincubated with 5 mA/ NaN.i and 50 mM 2-deoxyglucose for 30 min at 37°C
before incubation with vesicles. Ti)e metal)olic inhil)itors w(>r(> also pr(>sent during vesicle-cell
incubation.s.

DEPARTMENT OF EMBRYOLOGY

47

TABLE 5. Rcloaso of Coll-Asso(;ial,o(l DPL and
DML Vesicles from Colls by Trypsinization*

Tern I). % of Vesicl(>

Vesicle of Vesicle Radioa(;tivify

Typo Incubation (°C) Released from ('ells

DML

2

40-50

DML

15

29-42

DML

32

5-12

DML

37

1- 2

DPL

2

55-62

DPL

20

49-57

DPL

37

3- 9

* E-cells were incubated with unilamellar
lipid vesicles for 1 hr at indicated temperature,
then washed and treated with 0.01% trypsin for
10 min at 22°C. Measurements represent the
range of values found in four experiments.

and supernatant revealed that varying
amounts (^0-60%) of the cell-associ-
ated DPL or DML can be released into
the medium by trypsin, depending on the
temperature at which the initial vesicle-
cell incubation is carried out (Table 5).
Prolonged incubation with trypsin (up
to 30 min) does not further reduce the
amounts of cell-associated DML or DPL.
All of the radioactivity released by
trypsin is chloroform-methanol extract-
able and chromatographs as lecithin.
Spontaneous release of ^"^H-DPL or DML
from cells in the absence of trypsin is
negligible. The release of DPL or DML
cannot be induced with heat-inactivated
trypsin or with trypsin inhibitor alone.
Incubation of cells with DML or DPL
vesicles has a significant effect on the
susceptibility of certain cell-surface pro-
teins to lactoperoxidase-catalyzed iodi-
nation. Figure 31 A-C presents auto-
radiograms of the SDS-polyacrylamide
(7.5%) gel electrophoresis patterns ob-
tained for the proteins from ^^^I-labeled
control cells (Fig. 31 A), and for cells
preincubated 1 hour at 2°C with DPL
(Fig. 31 B) or DML (Fig. 31 C) vesicles.
While many of the radiolabeled bands
are unaffected by vesicle treatments, the
extent of ^^^I-labeling in several regions
(I-V; Fig. 31) of the gel was obviously

different for control than for DI^L or
DML vesicle treatments. A prominent
difference is in region III of the gels, in
which it is seen that two bands, cor-
responding to a molecular weight of
about 60,000 daltons, are heavily labeled
in control cells, and apparently unlabeled
in the vesicle-treated cases. In contra.st,
the labeling in regions I, II, IV, and V
of the gels aj)i)ears to be enhanced for
DPL and DML treatments, compared to
the control. Coomassie l)lue staining pat-
terns of the gels corresponding to Fig. 31
A-C were virtually identical.

EDTA-dissociated cells were also in-
cubated with '"^H-DML vesicles 1 hour at
2°C, and the cellular proteins from
washed cells were solubilized with SDS
and subjected to poly aery lamide gel
electrophoresis. Figure 31 E,G shows the
fluorographic patterns obtained in this
experiment using 7.5% and 15% poly-
acrylamide gels. While most of the -^H
radioactivity migrates at the front of
the gel, presumably as SDS-DML mi-
celles, a number of radioactive bands
also appear throughout the gel. It is note-
worthy that two bands (Fig. 31 E,G,
arrows) are heavily labeled with tritium,
and that the position of these bands cor-
responds to the position of the two pro-
tein bands whose ^^'""I labeling was in-
hibited by vesicle treatments (Fig. 31
B,C). All of the '"^H-cpm in these bands
is chloroform-methanol extractable. More
than 75% of the radioactivity chromato-
graphs as lecithin, the rest as lysolecithin
and free fatty acid.

Fig. 32 A-D shows a typical scanning
electron micrograph of an EDTA-dis-
sociated cell in the absence of vesicle
treatment. Numerous projections and
attachments to the glass substrate are
visible (Fig. 32 A, arrows) 5 minutes
after plating of the cells. Microvilli also
appear to be extended away from the
cell body. At high magnification the re-
gions of the cell surface between projec-
tions appear smooth (Fig. 32 B). In
contrast, cells treated with DPL vesicles
for 1 hour at 2°C differ from untreated

48

CARNEGIE INSTITUTION

moi wt
X 10"^

moi wt
X 10"^

130

68

57.5

43.5

V

I '
1'

in

68

57.5
43.5

27

3.4

A B C D

F

G

H

Fig. 31. SDS-polyacrylamide gel electrophoresis patterns obtained from control and vesicle-
treated cells. Autoradiograms are for the ^^I-labeled proteins from (A) EDTA-dissociated
(E-celLs), (B) DPL-treated E-cells, (C) DML-treated E-cells, and (D) trypsin-treated
(T-cells). Fluorograms are for (E, G) E-cells and (F, H) T-cells treated with 'H DML vesicles.
All vesicle treatments were 1 hr at 2°C. (A-F) 7.5% gels; (G, H) 15% gels. The positions of
the molecular weight markers, /3-galactosidase (130,000); BSA (68,000); catalase (57,000);
ovalbumin (43,500); ConA (27,000); and cytochrome C (13,400) are shown.

cells in several respects. Few microvilli
extend outward from the cell surface into
the bathing medium (Fig. 32 C) ; instead
they appear to be held lengthwise on the
cell body. Furthermore, no cellular pro-
jections extending onto the glass cover-
slip are seen. At higher magnification
(Fig. 32 D) the cell surface has a rough
or bumpy appearance compared to con-
trols (Fig. 32 B).

In separate studies, cells were incu-
batf'd with DPL vesicles containing a
trapped, water-soluble fluorescent dye, 6-
carboxyfluorescein (6-CF), and subse-
quently examined by fluorescence mi-
croscopy. When such incubations were

carried out at low temperatures, the
vesicle-treated cells were seen to be sur-
rounded by a bright ring of fluorescence
(Fig. 33 A), while incubations carried
out at 37°C resulted in an even distribu-
tion of dye throughout the cell (Fig.
33 B). Very little fluorescence is seen in
control populations of cells treated with
DPL vesicles containing no 6-CF or with
cells incubated with a solution of vesicles
and untrapped dye (not shown).

The interactions of DML and DPL
vesicles with trypsinized (10' at 37°C)
cells were also studied. Depending on the
temperature of vesicle-cell incubation,
these cells incorporate about L5-2.5

DEPARTMENT OF EMBRYOLOGY

49

Fig, 32. Scanning electron micrographs of EDTA-dissociated cells. (A, B) untreated controls;
(C, D) DPL-treated (1 hr at 2°C). Arrows in A indicate typical attachment sites to glass
substratum. (E) Trypsin-treated cell incubated (1 hr at 2°C) with DPL vesicles. Several
spherical structures (/->-' 300- 1000 A diam.) resulting from vesicle treatment are shown at arrows
in E. Low magnification, 7000 X ; high magnification, 28,000 X-

times less DML or DPL than nontryp- treatments. Furthermore, the two protein

sinized cells (Table 4). The iodination bands (Fig, 31 E,G) which are heavily

pattern in trypsin-treated cells (Fig. 31 labeled with ^H DML in EDTA-disso-

D) is not significantly affected by vesicle ciated cells, are only lightly labeled (Fig.

50

CARNEGIE INSTITUTION

Fig. 33. Fluorescence micrographs of E-cells treated with DPL vesicles containing 6-carboxy-
fluorescein for 30 min. (A) E-cells, 2X; (B) E-cells, 37°C.

31 F. H) in trypsin-treated cells. Scan-
ning electron micrographs of trypsin-
treated cells were qualitatively similar to
those obtained with EDTA-dissociated
cells except that the number of particles
adsorbed to the cell surface was obviously
less fFig. 31 E vs. 31 D).

Temperature Dependence of
Vesicle Uptake

The temperature dependence of DML
and DPL vesicle uptake by EDTA-
dissociated and trypsin-treated cells is
shown in Fig. 34. Cells were incubated
for 1 hour at the indicated temperature,
washed, and then assayed for the uptake
of exogenous lecithin. For EDTA-disso-
ciated cells, DML uptake is nearly con-
stant between 20 and 40°C but increases
markedly below about 19-20°C. For
DPL. vesicle uptake increases with de-
creasing temperatures over the entire
range of temperatures examined. The
rate of increase is greater, however, be-
tween 40 and 30 °C than between 30 and
0°C. Treatment of cells with trypsin (10
min, ST'^C) prior to incubation with vesi-
cles .significantly alters the temperature
dependence of vesicle uptake. For DML,
the uptake between 0"^ and 39 ^C becomes
nearly temperature independent. For
DPL, trypsinization markedly reduces

vesicle uptake at low temperatures, re-
sulting in an uptake profile with little
temperature dependence. Prolonged tryp-
sinization of cells (up to 30 min) does
not further modify vesicle uptake.

Influence of Tc on Exogenous Lipid
Incorporation by Cells in Suspension

Although this report deals primarily
with the adhesion pathway for vesicle
uptake found at temperatures below the
Tr of DML or DPL, it should be noted
that above this temperature, there is a
different mechanism of uptake. Thus,
significant amounts of cell-associated
DML or DPL can be released by trypsin
when vesicle-cell incubations are carried
out below the vesicle T,., but only negli-
gible amounts are released after incuba-
tions above this temperature (Table 5).
Based on our previous studies with V79
cells in which we used ''fluid" EYL or
DDL vesicles, it seems likely that the
mechanism of DML and DPL uptake
above their T,. involves vesicle fusion
with the cell surface. The finding of a
uniform fluorescence in cells treated at
37°C with DPL vesicles containing 6-CF
(Fig. 33) is also consistent with a fusion
mechanism above the vesicle Tr.

The uptake of DML or DPL vesicles
is not the result of a significant contri-

DEPARTMENT OF EMBRYOLOGY
30

51

UJ

<

EDTA DISSOCIATED

10

20

TEMP rc )

30

40

EDTA DISSOCIATED

10

20
TEMP (*»C )

30

40

Fig. 34. Temperature dependence of vesicle uptake by EDTA-dissociated and trvpsin-treated
cells. (A) DML and (B) DPL.

bution of an endocytotic pathway, since
saturation levels of uptake were reached
after only 5-10 minutes of incubation at
37°C (data not shown). If vesicle uptake
involves an active process, then incuba-
tion for times greater than 5 minutes
should increase the uptake. Furthermore,
involvement of an active cellular process

requires a significant increase in uptake
with increasing temperature. However,
in every case, DML or DPL uptake was
either greater at lower temperatures or
showed no significant temperature effect
(Fig. 34) . Finally, treatment of cells with
combined inhibitors of respiration and
glycolysis (Table 4) has virtually no

52

CARNEGIE INSTITUTION

effect on DPL uptake and results in only
a small inhibition i-2CK(0 of DML up-
take at 37'C.

The findings reported here support a
vesicle-to-cell adsorption process at low
temperature in both EDTA-dissociated
and trypsin-treated cells. Scanning elec-
tron micrographs of V79 cells (Fig. 32
D and E) reveal a characteristic bumpy
surface following vesicle treatment. The
size oi such "bumps" ranges from about
300 to lOOOA, and is consistent with the
adsorption of DPL or DML vesicles to
the cell surface. DPL treatments of cells
seem to hold microvilli to the cell body
along their extended length, which pre-
vent;? their full extension into the bathing
medium and inhibits spreading of the
cells onto a glass substrate. This could
be explained if the adsorbed vesicles act
as a bridge, binding different regions of
the cell surface together. EM auto-
radiosrams (Year Book 75) showed an
accumulation (^90% of ^H DML or ^^H
DPL at the cell periphery after a low-
temperature incubation, suggesting that
the adsorbed particles seen in Fig. 32
represent applied vesicles. Finally, fluo-
rescence microscopy of cells treated with
vesicles containing the water soluble
fluorescent dye 6-CF show an intense
rinfi of fluorescence at their ]:)eriphery
(Fig. 33 A). Such a pattern of fluores-
cence is additional evidence that vesicle-
to-cell adsorption is the dominant path-
way of uptake at temperatures below the
vesicle Tr.

Mode of Veside-to-Cell Adhesion

Our data suggest the importance of
cell surface proteins in the vesicle-to-ccll
adsorption phenomenon discussed above.
The finding that uptake of vesicles by
EDTA-dissociated cells is always greater
than that by its trypsinized counterpart
suggests that a trypsin-sensitive material
is required for the greater })inding of
vesicles to the cell surface. The release
of a significant amount of DPTv or DML
from the cell surface by trypsin (Table

5) is also consistent with the involve-
ment of a cell-surface protein moiety in
vesicle binding. Scanning electron micro-
graphs of cells (Fig. 32 D and E) show
that the number of vesicles bound to
the non-microvilli regions of EDTA-
dissociated cell surfaces is much greater
than in trypsinized cells. This suggests
that the lower uptake of vesicles in
trypsin-treated cells is not simply due
to a reduction in the cell-surface area
available for vesicle-to-cell adhesion,
which could arise from subtle differences
in the diameter and number of microvilli
in the two cell types.

DML or DPL vesicle pretreatments of
EDTA-dissociated cells significantly
modified the lactoperoxidase-catalyzed
iodination of some of the cell-surface
proteins. The modifications in gel pat-
terns shown in Fig. 31 could result from
(1) altered mobility of proteins in the
gel due to the presence of tightly bound
exogenously supj^lied DML or DPL
lipids that were not completely removed
from the cellular proteins by SDS solu-
bilization and/or (2) modified accesi-
bility of some of the cell-surface proteins
to ^-^I-labeling by lactoperoxidase. While
we cannot completely exclude the first
possibility at this time, the absence of
heavily labeled ^H DML bands in re-
gions I, II, IV, and V of the fluorogram
shown in Fig. 31 E suggests that the en-
hanced iodinated bands seen in those
regions following vesicle treatment (Fig.
31A, B, C) are not due to an anomalous
migration of proteins in gels resulting
from their incomplete delipidation. Our
finding of an apparent inhibition of ^^^I
labeling of the approximately 60,000
MW cell-surface proteins in the presence
of adherent vesicles (Fig. 31 A vs. 31
B, C) is consistent with explanation (2).
Thus, gel electrophoresis and fluoro-
graphy showed that the '"^H DML, al-
though associated with a num})er of
protein bands, was strongly associated
with the two proteins whose iodination
is inhi})ite(l })y vesicle pretreatments of
cells (arrows, Fig. 31 E, O) . Furthermore,

DEPARTMENT OF EMBRYOLOGY

53

this association was considerably weak-
cned or nearly absent (Fi^. 31 F, H) in
trypsin-treated cells, which also exhibit
a greatly reduced DML or DPT^ vesicle-
to-cell adhesion (Table 4; Fig. 32 D, E).
Thus, we conclude that vesicle })inding
to the surface of EDTA-dissociated cells
involves a number of cell surface pro-
teins, several of which have a high affin-
ity for vesicle lipid and are protected
from iodination as a result of vesicle
binding.

The adsorption of DML and DPL
vesicles to cells may also involve a
trypsin-insensitive protein, or a non-
protein component of the cell surface.
We suggest that vesicle adhesion to
trypsin-treated cells may be primarily
to such nonprotein regions of the cell
surface, while adhesion to EDTA-dissoci-
ated cells involves both protein and non-
protein elements. In agreement with this
idea is our finding that DML or DPL
vesicle treatments of cells affects the
iodination pattern in EDTA-dissociated
but not trypsinized cells. Since the bind-
ing of DML or DPL vesicles to EDTA-
dissociated cells at low temperatures is
much greater than to trypsin-treated
cells, adsorption to surface proteins prob-
ably dominates the uptake process at this
temperature. With increasing tempera-
tures, the level of vesicle uptake by the
two cell types approached one another
(Fig. 34), suggesting that adsorption to
the nonprotein part prevails. We specu-
late that this nonprotein part of the cell
surface involved in vesicle uptake may
represent accessible regions of plasma
membrane lipids.

In the present study we showed that
below their T,, DML and DPL lipid
vesicles adhere mostly to a trypsin-
sensitive material on the surface of
EDTA-dissociated cells. Our findings
suggest a possible pathway for inter-
cellular adhesion involving both protein
and lipid. According to such a scheme,
trypsin-sensitive materials on one cell
surface, which have the ability to com-
bine with some of the exposed lipids from

the plasmalemma of another cell, result
in th(! adhesion of the two surfaces. By
analogy to our finrling that only '^^;olid"
vesick.'s such as DML or DPL below
tlunr Tr, form stable adhr^sions to cells,
it is suggested that the lipids in those
regions of plasma membrane involved in
cell-to-cell adhesions are highly spe-
cialized, pro})ably containing high pro-
portions of "rigid" or saturated acyl
chains compared to the plasma mem-
brane as a whole. This model differs from
current ideas on the molecular basis for
intercellular adhesion of cells, which
emphasize either specific interactions be-
tween cell-surface proteins and/or carbo-
hydrates of contacting cells, or inter-
actions between the plasma membrane
lipid bilayers of adjacent cells. While the
proposed model is not intended to explain
specific interactions between cells, the
present findings do suggest that lipid and
protein interactions between contacting
cells may be important in stabilizing
specific membrane contacts, once formed.

Binding and Capping of Fluorescent

Dye Containing Lipid Vesicles

TO Lymphocytes

K. Ozato, L. Huang, and R. Pagano

In Year Book 75 we demonstrated that
phospholipid vesicles are taken up by
murine lymphocytes predominantly by
vesicle-cell fusion or vesicle-cell adsorp-
tion mechanisms, depending on the mo-
lecular composition of the exogenously
supplied lipids. Here we further examine
the uptake of DPL and DOL (dioleoyl
lecithin) vesicles by lymphocytes using
fluorescence microscopy. With this
method living cells can be examined im-
mediately after vesicle treatment.

Freshly isolated mouse thymocytes
from CBA males 6-8 weeks old were
used in all experiments. DOL and DPL
vesicles were prepared by sonication in
the presence of a w^ater soluble fluorescent
dye, 6-carboxyfluorescein (6-CF). The
sonicated vesicles containing trapped dye

54

CARNEGIE INSTITUTION

were then separated from free dye by
chromatography on Sephadex G-25 and
immediately used. Thymocytes were in-
cubated at 2'"" or 37 ''C for 1 hour (1 mg
vesicle lipid ml; 4 X 10^' cells ^ml) with
6-CF containing vesicles washed three
times in balanced salt solution and ex-
amined in a Zeiss fluorescence micro-
scope utilizing epi-illumination.

Typical fluorescence micrographs are
shown in Fig. 35. Cells treated at 37°C
with DOL vesicles containing 6-CF
showed a relatively uniform and diffuse
distribution of dye throughout the entire
cell volume (Fig. 35 A), while cells in-
cubated with DOL vesicles at 2°C
showed very little dye uptake and could
not be photographed. Using other types
of fluid vesicles (such as egg yolk leci-
thin) containing trapped 6-CF, we ob-
tained results quantitatively similar to
those obtained with DOL. On the other
hand, cells treated at 2°C with DPL
vesicles containing 6-CF showed an ac-
cumulation of fluorescence at the periph-
er>' of cells and very little fluorescence
in the interior. The fluorescent vesicles
also appeared to form patches and caps
at the cell surface (arrows, Fig. 35 B).
When the incubation with 6-CF con-
taining DPL vesicles was carried out at
37°C (Fig. 35 C), greater fluorescence
intensity was observed at the cell bound-
ary, with some fluorescence inside the
cells. The ring of fluorescence was heavy
and extended from the cell surface in a
characteristically spiky fashion. No de-
tectable fluorescence was seen in cells
treated with vesicles containing no 6-CF,
or with vesicles plus untrapped dye.

The accumulation at the cell periphery
of 6-CF entrapped in DPL vesicles after
ve.^icle-cell incubation confirms the stable
ad.'^orption of this vesicle type to the
lymphocyte cell surface, which we re-
ported earlier. Peripheral localization of
the entrapped dye was accentuated when
ihf incubation was carried out at 37°C
(Fig. 35 C), suggesting a possible mul-
tiple adsorption of DPL vesicles. When
cells were incubated with dye containing

Fig. 35. Fluorescence micrographs of mouse
thymocytes treated with lecithin vesicles con-
taining 6-carboxyfluorescein (6-CF). Treat-
ments are (A) DOL vesicles (1 hr at 37°C) ;
(B) DPL vesicles (1 hr at 2°C) ; prominent
capping and patching of fluorescent vesicles at
the cell periphery indicated by arrows; and (C)
DPL vesicles (1 hr at 37°C) ; fluorescent ring
was often non-uniform and extended away from
cells (as shown by arrows). (B) and (C) are
composite photographs from two different fields.
Bar is 5 Ai.

DPL vesicles at 2°C, patching and cap-
ping of the adsorbed vesicles at the cell
surface was evident (Fig. 35 B). This
phenomenon might be due to the subse-

DEPARTMENT OF EMBRYOLOGY 55

quent warming of samples on the micro- We are investigating the possibility

scope stage during observation, even that this reorganization of intact, adher-

though it was observed immediately upon ent vesicles on the lymphocyte surface

examining the cells in the fluorescence represents a reorganization of cell-surface

microscope. lipids or proteins, or both.

REGENERATION IN THE NERVOUS SYSTEM:

FORMATION OF SPECIFIC SYNAPSES IN THE LEECH

K. J. Muller and S. T. Carbonetto
with the technical assistance of B. Thomas

A central problem in neurobiology is: devised a microscopic histochemical tech-
How do growing neurons select the nique for marking single leech neurons
targets with which they synapse? During through recording microelectrodes. As
development, when the nervous system was described in Year Book 74 and Year
is being wired together, a single neuron Book 75, we inject a tissue sample with
might synapse with just a few cells from the enzyme horseradish peroxidase
a pool of perhaps thousands or tens of (HRP) under pressure through a micro-
thousands. Regenerating neurons in the pipette and stain the sample to detect the
central nervous system of adult verte- enzyme. In the leech both types of
brates can also establish specific con- neuronal synapses — chemical and elec-
nections. However, to show that precisely trical — are readily detected physiologi-
the same neurons are reconnecting — that cally with microelectrodes. Ordinarily
is, to study regeneration at the level of synapses can be identified morphologi-
single cells — it has been necessary to cally only in the electron microscope, but
turn to simpler systems. The nervous we reported that with the HRP technique
system of the leech has been especially the synapses of specific sensory and
useful for this, since it has only a few motor cells could be localized in regions
hundred neurons in each segmental gan- of the cell that are distinctive in the
glion and the functions of many of the light microscope. Another feature of the
cells have been identified. The synaptic HRP technique — that the enzyme does
interactions between these neurons have not pass between electrically coupled
been found to be surprisingly consistent neurons— was used to demonstrate the
and are, in that sense, predictable from precise location of the single electrical
one ganglion or animal to the next. Con- synapse between pairs of identified inter-
sequently, it has been possible to show neurons (the Rohde, or S neurons) . :\Iuch
with electrophysiological techniques that ?^ our work durmg the past year has
1 • J.' • 1- '1 ^ focused on regeneration of this particular
during regeneration individual sensory ^-rr i , i • i ^ -i
1 , ,1 . 1 , synapse. We have traced m some detail
neurons can select their normal post- J . ,1 i xi x-

^ , , 1 1 1 the steps taken by the regenerating

synaptic contacts from among hundreds ,-, \ i^i- u ^-l ■ ^ 4. ■

, ,i ,. , , , .1 .1 neurons as they reestablish their electri-

of other candidates, and re-form chemical ^^j connections ; and we have studied the

synaptic connections. Still, we have very ^^^^ ^^ ^^^ ^^^^^^ -^ re-forming synapses

little information on the structure of ^^^ ^q^, ^^^ n^,,. synapses compare in

regenerated neurons and their synapses structure and distribution to the old.

or on how they go about finding their ^g fi^^^ two functionally equivalent but

postsynaptic targets. structurally distinct mechanisms for the

With the aim first of detailing the regeneration process, and there is e^d-

structure and distribution of synapses in dence that each has its counterpart in

physiologically well-studied neurons, we other nervous systems.

56

CARNEGIE INSTITUTION

Regexeratiox of ax Electrical Synapse of these structures that are typical of

gap junctions.
A'. J. MJlcr, S. T. Carbonetto. and B. Thorrias

The Electrical Synapse between S-cells

To study experimentally the formation
of synapses between specific neurons, it
is necessary to recognize the elements
involved and, for comparison, to have
an accurate picture of normal synaptic
connections. The electrical synapse be-
tween S-cells in the leech is particularly
useful for this. Other investigators have
shown that these neurons, thought to be
syncytial (hence ''S-cell"). could reestab-
lish electrical continuity after surgical
disruption of a connecting axon (Frank
et al, J. Comp. Xeurol. 159, 1, 1975).
Last year we reported that such electri-
cally connected S-cell interneurons, one
in each of the 21 segmental ganglia of
the animal, are linked at the ends of
their axons by electrical synapses. Except
near the extreme ends of the ventral
nerve cord, the axons of S-cells meet at
about the midpoint along the connectives
between ganglia. This single region of
contact between neighboring S-cells is
the site of the electrical synapse that
mediates passage of nerve impulses in
either direction as they travel the entire
length of the intact nerve cord.

In thin sections viewed in the electron
microscope, the presumed electrical con-
tacts or gap junctions between S-cells
resemble those of other invertebrates.
The synaptic membranes at the electrical
junction are separated uniformly by 6
or 7 nm and can have linear dimensions
of a micron. The synapsing axons con-
tact each other at the ends of several fine
processes rather than along an extensive
septum, perhaps to allow for modulation
of synaptic transmission, as discussed
in Year Book 75. To define better the
structure of the junction, we infused
fixed preparations with La + + + salts,
which fill extracellular spaces and thereby
negatively stain the gap junctions. We
found, rather than a regular matrix of
20-nm hexagonal units, a scattered array

Eegeneration of the S-cell Axon

During development the S-cell axons
travel halfway to the next ganglion down
the connective and synapse with the
neighboring S-cell. We crushed the nerve
cord close to one ganglion (8, 9, 10, or
11) so that the electrical synapse itself
was not directly affected by the lesion,
and observed the course of regeneration
of this highly specific and localized con-
tact. This procedure severs all axons in
the bundle of nearly 100 (Faivre's nerve)
in which the S-cell axon travels. Days or
weeks after the lesion was made, the
nerve cord was removed to a recording
chamber, electrodes were placed into S-
cells on both sides of the crush, and the
degree of electrical coupling — if any —
was measured. Both cells were then in-
jected with HRP and processed for light
or electron microscopy.

It commonly takes 3 weeks for elec-
trical coupling to resume between S-cells.
However, anatomical studies show that
within a week after the lesion a spray of
sprouts emerges from the end of the
severed axon. At the crush the sprouts
are not restricted to Faivre's nerve, but
beyond the crush one finds only one or a
few fine branches that course along the
route of the old axon, as if sprouts that
find the former pathway are stimulated
to continue growing. By contrast, the
intact S-cell axon at the other end of the
connective (the ''target" S-cell), its prin-
cipal input removed, does not sprout but
instead awaits reinnervation by the in-
jured neuron (Fig. 36 A).

Sections through the connective con-
firm that the regenerating neuron closely
follows its severed distal stump after
crossing the lesion (Fig, 36 B). The
severed distal axon, which looks almost
normal, is longitudinally indented with
grooves in which the fine regenerating
sprouts are nestled. A glial sheet wraps
both the old and the new axons and

DEPARTMENT OF EMBRYOLOGY

57

l8d

crush

I

0.2 mm

B

Fig. 36. S-cell axons regenerate toward synapse with target S-neuron. Both cells were injected
with HRP 18 days after the connective was crushed. (A) Ventral view of connective linking
ganglia. Regenerating axons (top) grow across crush and down middle of connective toward
thicker and darker target S-cell axon. (B) Portion of connective cross section at upper arrow
in A, after osmication and embedding in Epon. Densely stained growing tips of regenerating
neuron course along the severed distal stump, which has as usual a large and light-staining
profile.

separates them everywhere except near
the growing axon tips and in small re-
gions along their length. Throughout the
3- or 4-week period before electrical
coupling is restored, while the two S-cells
have not yet met, the severed distal
stump persists as a healthy-looking and
(as will be shown below) functional
axon.

Reformation of the Synapse

Typically the next stage — restoration
of electrical coupling — occurs within 4
weeks, followed rapidly by transmission
between adjacent S-cells. A consistent
and remarkable finding, however, is that

in the earliest stages impulses are con-
ducted only from the target S-cell into
the regenerating neuron, not in the re-
verse direction (Fig. 37). Such one-way
transmission between S-cells might in
principle be due to a rectifying electrical
junction such as occurs between certain
sensory and motor neurons in the leech,
but current passed into both cells demon-
strates that they are reciprocally coupled.
Moreover, it is sometimes possible to
overcome the unidirectional block of
transmission by depolarizing one S-cell
with a suitably timed current pulse.
Taken together with the small caliber
of the regenerated process at one month,
the unidirectional transmission of im-

58

CARNEGIE INSTITUTION

30 d

20msec

0.1 sec

Fig. 37. Impulse transmission across newly formed synapse between S-cells is at first one-way.
Connective was cruslied near ganglion 8. 30 days before. S-cells were impaled with micro-
electrodes, and a suction electrode applied to the posterior (/)) connective. Impulses generated
in ganglion 9 or more posteriorly will invade the regenerating S-cell in ganglion 8, but not vice
versa. Because currents pass well in both directions across the synapse, the junction is non-
rectifying. Both cells were later injected with HRP and stained.

pulses is consistent with an inability of
the regenerating neuron to deliver enough
current to the target S-cell to excite it.

When the coupling of neighboring S-
cells can first be recorded, HRP injec-
tions show that the cells' axons are con-
tiguous. Where the target S-cell and the
distal stump overlap and contact each
other, the regenerating neuron interposes
itself between them and branches out to
associate with the finer processes of the
target S-cell (Fig. 38). The regenerating
neuron forms electrical synapses here
and docs not grow beyond the normal
region of synapse. During the next month
the caliber of the regenerated axon in-
creases; full coupling and transmission
are restored and the distal stump rapidly
degenerates.

It seems likely that the formation of
the new synapse, which may involve a
displacement of the old, triggers the dis-
appearance of the distal stump. If re-
generation fails (20 to 30% of total)
the distal stump persists for 5 or more
months and slowly degenerates. This de-
generation is marked by glial hyper-
trophy and a shrinking and darkening of
the S-ceH axon, which is still recognizable
by its distinctive position in the nerve.

Once impulse conduction is established
and the functional axons begin to in-
crease in caliber, thereby increasing the
speed and reliability of transmission, the
extra sprouts at the crush disappear.
This chain of events is typical of re-
generating nerves and confirms the suit-
ability of the S-cell as a model for
synapse formation.

The sequence of regeneration can be
considered in steps. The initial growth is
not well directed, but once contact is
made with the old pathway, highly
directed growth proceeds. When synapses
have formed with the target, growth
stops, and the distal stump begins to
degenerate. We do not yet know the
nature of the intercellular triggers for
the progression of these stages, but the
system for the assay is now in hand.

''Fusion''

The six cases of most rapid restora-
tion of coupling between S-cell pairs
(from 10 days to 3 weeks) provided the
evidence for another mechanism by which
S-c(»l]s can reconnect. The regenerating
S-cell had not reached the target S-cell,
as shown })y HRP injection, so coupling

DEPARTMENT OF EMBRYOLOGY

59

V

,^,/ }

. ^

-'i l'"'/''"

31 d

y

^^u\

Fig. 38. Regenerating neuron (*) interposed between target S-cell (stained axon profile on
left) and severed distal stump (right). Same degree of regeneration as in cell in Figure (37).
Branches of the regenerating cell (*) have followed the stump's branches to contact the target.

was through an intermediate element
which we now know is the distal stump.
The regenerating axon did not course as
one or two fine fibers along the distal
stump, but instead formed around the
severed axon a distinctive basket of

broader processes which could be unam-
biguously identified in either the light
or the electron microscope by its large
size, dorsal position in Faivre's nerve, and
lightly staining cytoplasm containing
scattered mitochondria (Fig. 39). Sheets

60

CARNEGIE INSTITUTION

jCXUS

Fig. 39. When the S-cell forms an electrical synapse on its own severed distal stump, the
regenerating processes surround the stump. (A) The axon proximal to the crush (below)
branches as it crosses the crush in the center of the connective and remains branched as it
forms a basket of proce.sses along the site of the distal stump. The marked cell ends abruptly
several hundred micrometers before reaching the next S-cell (not shown), which had also been
stained. (B) The growing tip. In this plane of focus, the marked processes are seen to surround
portions (indicated at three locations by arrows) of the distal stump (18 days). (C) Schematic
diagrams from peroxidase-injected preparations indicating the normal and experimentally
induced connection (anterior at top). (D) Electron micrograph of connective cross section
showing the two stained processes of the regenerating neuron (i^i and R2) near the distal stump
(5), identified by its large size, dorsal portion in Faivre's nerve, and hght cytoplasm. Inset:
where the regenerating axon (R^) apposes the segment are points at which the membranes are
separated by no more than 4-5 nm (arrowheads), the presumed gap junctions (34 days).

of glia separated much of the area be-
tween the regenerating neuron and its
severed axon segment, but direct contact
was made near the growing axon tip and
in some other regions. These small areas
of close apposition appear to be gap
junctions and are the presumed sites of
electrical coupling. Tilting the thin sec-
tions in the electron microscope confirms
that the membranes are separated only
by some 4 to 6 nm.

For there to be electrical coupling and
transmission through the severed axon
stump, the axon stump must survive
physiologically and continue to excite
and be excited by the uninjured target
S-cell across the old electrical synapse
during the month after the connective is
crushed. We demonstrated this by stim-
ulating and recording from the distal

stump in a preparation that had not
regenerated (Fig. 40).

Coupling and subsequent transmission,
first in only one direction, occurs in the
same sequence through the distal stump
as when S-cells contact each other di-
rectly (Fig. 41 A and B). It seems likely
that the mechanisms for one-way trans-
mission are the same.

The electrical synapse formed by the
regenerating neuron on its own distal
stump is functionally indistinguishable
from a direct fusion with the stump, ex-
cept that large molecules such as HRP
(40,000 daltons) are unable to cross the
junction. This raises the possibility that
for the many other systems in which
fusion has been suggested as the expla-
nation for rapid and complete regenera-
tion, an electrical synapse has also

DEPARTMENT OF EMBRYOLOGY

61

stim. p

r

stim. S

10

stim. a

f

0.5 mV I

0.5mv|

10 msec

Fig. 40. The distal axon stump (here of S-cell in ganglion 11) remains functionally connected
to the adjacent target S-cell (in ganglion 10) after crushing close to ganglion 11 (19 days previ-
ously). Suction electrodes record anteriorly (a) and posteriorly (p) to ganglion 10, in which
the target S-cell (Sw) has been penetrated with a recording microelectrode. Prior recording and
stimulation posterior to the crush (=) demonstrated that the S-cell connection from ganglion
11 had not regenerated. The recordings here illustrate respectively that (stim. p) threshold
stimulation at p excites Sw, whose impulse is also recorded at a; (stim. >Sio) direct activation of
Sio with a depolarizing pulse of current through the recording microelectrode activates not
only its own axon recorded at (a), but also the axon of the distal stump recorded at p; and
(stim. a) anterior connective stimulation at threshold directly activates Sw and subsequently
the distal stump at p. In the diagram on the left, the thickened line in the connectives repre-
sents the peroxidase-stained axon of >Sio, which terminated more than one millimeter from
point p. At this distance the suction electrode {p) would not have directly activated or
recorded from Sw.

formed. It is not yet clear what factors
determine whether the stump receives a
synapse from the regenerating neuron.

Our studies indicate that coupling
through the distal stump is an inter-
mediate step in a sequence leading to
direct contact between the regenerating
S-cell and its target S-cell at the site of
the old synapse. When coupling through
the distal stump is seen as late as a
month after the lesion, the distance be-
tween S-cells is as little as 100 /xm. At
later times no coupled, adjacent S-cells
have been seen that are not in direct
contact with each other. The distal stump
seems to be an imperfect target for the
regenerating neuron, since the neuron
continues growing to form its final
synapse.

Survival of the Distal Stump

From the data presented so far one
might conclude that the persistence of
the distal stump is crucial to the entire
regeneration process. The stump serves
as a guide that is recognized by the
growing axon, and the stump's continued
contact with the target S-cell may pre-
vent that cell from sprouting or retract-
ing. We have initiated a series of experi-
ments to determine by what means the
distal stump survives without a cell body
for weeks or months and whether the
target S-cell will sprout when the stump
degenerates.

One hypothesis is that the stump is
sustained by the surrounding glia or
through the electrical synapse. A single

62

CARNEGIE INSTITUTION

12d.

t='-4mw

0.1 sec

20d

20 msec

0.1 sec

Fig. 41. Stages of coupling through distal stump. (A) Earliest sign of coupling: Coupling is
too weak to sustain impulse transmission 12 days after crush near ganglion 8. Impulses in distal
stump, first generated in Si, are seen electrotonically in Ss. Junction is nonrectifying. (B) Before
the regenerating S-cell reaches its target S-cell, there is already two-way transmission through
the distal stump. The connective between ganglia 8 and 9 was crushed 20 days before. S-cells
(Ss and iSt.) that, on subsequent marking with peroxidase, were found not to have regenerated
a direct contact nevertheless show normal functioning. Upper: impulses generated in Ss with
the recording microelectrode are propogated into ^9 and vice versa. Lower: hyperpolarizing
currents injected into S^ or >Sb spread into the adjacent S-cell, indicating good coupling through
the uninjected process.

glial cell ensheathes one entire connective ably i)laced crush. We have examined a

and Faivre's nerve. Its nucleus lies mid- series of animals in which the S-cell axon

way between ganglia and can be sepa- was crushed in several places. Our pre-

rated from the distal segment by a suit- liminary results indicate that distal axon

DEPARTMENT OF EMBRYOLOGY

63

segments survive, though less well than
the entire stump, when just a portion of
axon is isolated. Moreover, S-cell re-
generation is not impeded along the iso-
lated segment. In other experiments,
crushing the connective at both ends
isolates both axons but does not initiate
degeneration.

Several months after regeneration has
been experimentally prevented, we have
observed a reorganization of S-cell mor-
phology. In one instance, after 6 months
without its normal connection the target
S-cell entirely withdrew the "unused"
axon from the connective.

Using culture techniques described last
year, we hope to achieve regeneration of
the S-cell axon and synapse in vitro.
This would permit considerable manipu-
lation of the system. Thus far, functional
restoration of transmission has been only
through polysynaptic pathways.

The DisTRinuTioN of Synapses of
Identifip:d Sensory Neurons

K. J. Muller and B. Thomas

We are continuing the investigation of
the formation and distribution of chemi-
cal synapses by mechanosensory neurons.
A study of branching patterns and
synaptic enlargements of homologous
neurons in successive ganglia of single
animals has provided an index of the
degree of variability in genetically, de-
velopmentally, and functionally identical
neurons. Our findings will be useful in a
study under way with Dr. Eduardo
Macagno at Columbia University, in
which a computer graphic display of
injected neurons is being used to pinpoint
intercellular contacts. These contacts can
later be identified as synapses with the
electron microscope. Such methods should
provide for a straightforward analysis of
regeneration of specific connections.

STRUCTURAL AND FUNCTIONAL STUDIES
OF THE FIBROIN GENE

Y. Suzuki, Y. Ohshima, P. Geshelin, Y. Tsujimoto, and P. E. Giza

Our long-range goal is to understand
fibroin gene control by reconstituting its
function in vitro with the purified gene
and necessary components. Last year we
concentrated our efforts on the cloning
of the fibroin gene, with its presumed
regulatory sequences. We obtained 20
clones of fibroin gene plasmids. Each
plasmid was amplified in E. coli, and the
fibroin gene is now available in mg
quantity. One plasmid, pFbl9, carries
about 6000 bases (6 kb) of the 5' end
coding sequence and 12 kb of a flanking
sequence adjacent to the 5' end of the
gene. Another, pFblO, contains 6 kb of
the 3' end coding sequence and L3 kb
of a flanking sequence adjacent to the
3' end of the gene. More recently we
have obtained a plasmid, pFb29, which
carries the entire coding sequence (17
kb) with the flanking sequences adjacent

to the 5' end (4.6 kb) and the 3' end
(less than 0.5 kb). Using these plasmids,
we have prepared restriction maps of the
gene. The site of a presumed primary
transcription of the gene has been mapped
at or near the site where the 5' end of
mature mRNi^ was mapped. We are now
studying the sequencing of the putative
regulatory regions as well as the spacer
sequences surrounding the gene.

Cloning of Fibroin Gene Plasmids

Y. Ohshima and Y. Suzuki
with technical assistance by P. E. Giza

In Year Book 75 (p. 37) we reported
our success in enriching the fibroin gene
about 40- fold. However, the average
molecular weight of this preparation was
found to be 10-12 X 10^ daltons, and it
was seriously nicked. We intended to

64

CARNEGIE INSTITUTION

clone the entire fibroin gene (11.4 X 10^
daltons^i with its flanking sequences by
the dA dT joining method, but since
nicked DNA is not satisfactory for the
tailing reaction by the terminal trans-
ferase, we abandoned this enriched
preparation.

Better DNA preparations were ob-
tained by isolating posterior silk gland
nuclei and extracting DXA from them
(Suzuki and Giza. 1976'i. When ex-
tracted, this DNA had a double-strand
molecular weight of 60 X 10*' daltons and
a single-strand molecular weight of 30 X
10*^^ daltons. indicating that single-strand
nicks were practically imdetectable. The
high molecular weight DNA was mildly
sheared in the presence of high salt down
to 25 X 10*^ daltons (double stranded)
and subjected to actinomycin D/CsCl
density gradient centrifugation (Fig. 42).
To our surprise, fibroin gene showed a
bimodal distribution in the gradient (Fig,
42 D ; one peak rej^resents intact genes
characteristic of the size class of the
bulk of DNA, and the other results from
gene fragments smaller than 10 X 10^
daltons. The bimodal distribution was
not due to the mild shearing because
even unsheared high molecular weight
DNA displayed the same proportion of
intact to fragmented genes as that shown
in Fig, 42 I. This observation might
mean that active genes in specific tissues
are more susceptible to nuclease attack
than inactive genes (Weintraub and
Groundine, Science 193, 848, 1976). Sepa-
rate experiments showed that the intact
size class of fibroin genes kept the same
single-strand molecular weight through-
out at least one cycle of actinomycin
D/CsCI centrifugation and DNA re-
covery from the gradient. Each size class
of the gene fshown in brackets in Fig.
42 1 1 was pooled separately and sub-
jected to the second cycle of actinomycin
D/CsCI centrifugation. The fragmented
gcno fraction (Fig. 42 TI })) was enriched
about 1000-fold, giving a gene concen-
tration of about 4%. The intact gene
fraction (Fig. 42 II aj was then sub-

jected to a third cycle of actinomycin
D/CsCl centrifugation (Fig. 42 III a)
and enriched about 15- fold overall (gene
concentration 0.06%).

Poly (dT) tails were added to the 1000-
fold enriched fraction, and the DNA was
annealed with poly (dA) -tailed pMB9
DNA, Although the efficiency of trans-
formation was very low (Table 6, column
2) in contrast to that by pMB9-dA/
p]\IB9-dT annealing mixtures (Table 6,
column 1 ) , we obtained about 230 tetra-
cycline-resistant transformants. From
these we found 12 fibroin gene clones by
^-^T-mRNA hybridization, at least 6 of
which were fotmd to be independent of
each other. The maximum insertion size
among these 6 clones was 4 X 10^ dal-
tons, and the average size about 2 X 10^
daltons. The largest one (pFblO) carried
6 kb of the mRNA coding sequence and
1,3 kb of a flanking sequence adjacent
to the 3' end of the gene, as described in
the next section.

We then tried to anneal the dT-tailed
high molecular weight gene fraction with
the dA-tailed p]\IB9 DNA, but the an-
nealing mixture did not contain any large
circles (below 1%). When a sheared bulk
DNA preparation from B. niori originally
without single-strand nicks was used as
test material for the tail addition, anneal-
ing, and transformation, the fraction of
circles and the efficiency of transforma-
tion were very low (Table 6, column 3) .
One reason for inefficient circle formation
and poor transformation was presumed
to be the occurrence of nicks in the DNA
which were newly introduced by a con-
taminating nuclease in the terminal
transferase and subsequent internal tail
addition during the terminal transferase
reaction. A preparation of terminal trans-
ferase obtained from W. Salser's labora-
tory gave a higher efficiency of trans-
formation (Tal)le 6, column 4), but the
fraction of circles in the annealing mix-
ture was still low. Other reasons for the
low efficiency of circularization and
transformation could be a low percentage
of tailed 3' ends among the total ends of

DEPARTMENT OF EMBRYOLOGY

65

<

5 10 15

Fract ion number

Fig. 42. Partial purification of fibroin gene by actinomycin D/CsCl gradient centrifugation.
The first cycle of actinomycin D/CsCI centrifugation (I) using 15 mg of Bombyx mori DXA
in 12 X 20 ml gradients. The gene-enriched fractions shown by the brackets (fractions no.
10-14 for Ila and no. 15-19 for lib) were subjected to the second cycle. Fractions no. 16-19 of
lib were used for cloning. Fractions no. 9-13 of Ila were subjected to the third cycle (Ilia).

Fractions no. 11-15 of Ilia were used for cloning. ( ), A260. ( — • — ), cts/min. An ahquot of

fractions was hybridized with ^^^-mRNA.

66

CARNEGIE INSTITUTION

TABLE 6. Summary of liA-dT Joining of the Fibroin Gene to the Plasmid pMB9

Experiment Xo.

Properties of DXA used
Purilication factor of

fibroin gene
No. average double strand

mol. wt. (XlO'daltons)
Wt. average single strand

mol. wt. (XlO'daltons)

Pretreatment of DXA

Terminal transferase
preparation used

Percent of circles on anneahng
with plMB9-dA

Tetracvcline-resistant cells/^ig
of pMB9-dADXAon
transformation

Fibroin gene clones/

tetracychne-resistant cells

Mol. wt. of largest fibroin gene

(PMB9-
EcoKl)

xiooo

XI

XI

xi

xi

Xl5

3.6

7

20

19

19

19

15

1.5

...

10-11

10-11

10-11

10-11

4

None

X'one

None

None

X-exo-
nuclease

APase

APase

Xo. 1*

Xo. 1

No.l

No. 2t

No. 2

No. 2

No. 2

15%

0-3%

2%
(3%)$

4-7%

(12-

21%)$

• • •

1 X 10' 1200 600 5800 2300 10300 12600

12/230 ... ... ... ... 7/16000

4X10' ... 11X10"

(2 X (4 X

10«)§ 10«)§

* Obtained originally from R. Ratliff .

t Obtained from W. Salser.

t Wt. % of circular structures.

§ Mol. wt. of average insertion.

(...) Either not apphcable or not tested.

the sheared B. mori DNA, or poor prim-
ing ability of many of the DNA ends.
We therefore tried treatment of the DNA
with A-exonuclease or alkaline phospha-
tase before the terminal transferase
reaction. The alkaline phosphatase treat-
ment increased the transformation effi-
ciency twofold and greatly improved the
weight fraction of the average molecular
weight of circular structures formed on
annealing (Table 6, column 6). The
A-exonuclease treatment was ineffective
(Table 6, column 5). Employing the
alkaline phosphatase treatment and the
better preparation of terminal trans-
ferase, we carried out transformation
with the dT-tailed high molecular weight
fibroin gene fraction annealed with dA-
tailf'd pMB9 DNA (Table 6, column 7).
Out of 16,(XX) tetracycline-resistant trans-
formants, we obtained 7 fibroin gene
clones, at least 6 of which were inde-

pendent of each other. The longest in-
sertion was 12 X 10^ daltons, and the
average insertion size was about 4 X 10^
daltons. The average size is much smaller
than the average size of the DNA used
(number average mol. wt. : 15 X 10^).
This result may be due to higher effi-
ciency of smaller DNA molecules in
forming circles or to nicks generated
during the purification procedure for this
specific preparation (Table 6, column 7).
The largest one (pFbl9) carries about
6 kb of the mRNA coding sequence and
12 kb of a flanking sequence adjacent to
the 5' end of the gene.

In summary, the dA/dT joining method
for cloning works well for cloning small
genes of less than 5 X 10^' daltons (7.5
kb). However, if the method is used to
clone a gene larger than 10 X 10^ daltons
and of low purity, one should use a nick-
free DNA and a nuclease-free terminal

DEPARTMENT OF EMBRYOLOGY

67

transferase. Restriction enzyme frag-
ments seem to be better for the terminal
transferase reaction, and when sheared
DNA is used the pretreatment with alka-
line phosphatase before tail addition may
improve the efficiency of cloning.

Recently a restriction map of fibroin
gene in bulk DNA of Bombyx mori has
been constructed by Manning and Gage
(Fed. Proc. 36, Abs., p. 878, 1977) and
by us. We now know that the restriction
enzyme EcoHl generates a DNA frag-
ment of 14 X 10^ daltons; it covers the
entire fibroin gene (messenger coding
sequence), about 4 kb of a flanking se-
quence outside the 5' end of the gene,
and possibly a trace (less than 0.5 kb)
of a flanking sequence outside the 3' end.
High molecular DNA (30-40 X 10^ dal-
tons) was extracted from the posterior silk
glands without fragmentation of fibroin
genes and digested with EcoRl, and the
fragment containing the gene was puri-
fied 60- fold by banding in an actinomycin
D/CsCl gradient and a sucrose gradient

sedimentation. From this DNA prepara-
tion we recently o})tained a clone carrying
the EcoKi fragment of 14 X 10'' daltons
which includes the whole gene ('pFb29j.

Characterization of Fibroin
Gene Plasmids

Y. Suzuki, Y. Ohshima, and P.E. Giza

Definite Identification of Fibroin Gene
in the Cloned Plasmids

Although the ^^^I-mRNA used for hy-
bridization has been judged to be more
than 95% pure, we could have cloned
DNA sequences other than those coding
for fibroin. Therefore, we wanted to iden-
tify the gene unequivocally in the plas-
mids. First, we worked out the hybridiza-
tion requirements for ^^^I-mRNA with
the gene plasmid pFblO at a variety of
mRNA concentrations (Fig. 43). As
much as 40% of the input mRNA hy-
bridized with the pFblO DNA (Fig. 43),
indicating that pFblO carries DNA cod-

I
O

X

c

E
\

I -

®

-

I / ®

x/

V

/ ^
/ \

f ^^-

' ^X~ -

1 1 1 ~~l~~-~-l--^ 1

- 40

Z3
Q.

C

20 o

200 400 600 800 1000

ng /ml

1200

Fig. 43. Saturation hybridization of '^'I-mRNA to pFblO. About 300 ng of pFblO, which
includes 75 ng of mRNA coding sequence, were used for each point. Up to 200 ng niRNA/ml
undiluted ^^'I-mRNA was used. For the points at 200, 400, and 1100 ng/ml (— O— ) the labeled
mRNA was diluted with cold mRNA, and taking the 200 ng/ml point as the standard, correc-
tions were made and plotted. ( — • — ), cts/min; (--x--), percent of input mRNA taken up
by the gene.

68

CARNEGIE INSTITUTION

CM

I

g

X

6

in
U

20 40

Fraction number

Fig. 44. The oligonucleotide profiles of RNase
Ti digests of '^'I-fibroin mRNA, and RNAs
hybridized with pFblO and pFbl9. (a) "^'I-
fibroin mRXA was digested by RNase Ti, and
the resulting oligonucleotides were fractionated
by a DEAE-Sephadex A25 column, (b) "^'I-
mRXA was hybridized with pFblO, and the
hybridized RNA was hberated by heating and
digested with RNase Ti. (c) The same as (b)
except that pFbl9 was used for the experiment.

ing for the major component (fibroin
mRNA) of the assay probe. Second, the
RNA that hybridized under saturation
conditions was recovered and digested
with RNase Ti, and the digest was frac-
tionated on a DEAE-Sephadex column.
The oligonucleotide patterns obtained
from pFblO and pFbl9 were indistin-
guishable from that of the input mRNA

(Fig. 44). This is the strongest evidence
that these plasmids contain the fibroin
gene. Third, a thermal denaturation pro-
file of pFblO was carried out (Fig. 45).
About 50% of the hyperchromicity
showed a sharp thermal transition at
high temperature, indicating that about
6 kb of the 12-kb plasmid represents a
high-GC sequence, as expected from the
known composition of the fibroin gene.
This evidence eliminates any ambiguity
about the nature of the cloned gene.

Screening and Restriction Mapping of

the Fibroin Gene Plasmids Carrying

Both mRNA Coding Sequence and

Flanking Sequence (s)

Manning and Gage {Fed. Froc. 36,
Abs., 1977) and the authors, working
independently, have prepared a restric-
tion map of the fibroin gene, using total
DNA from Bombyx mori. It was found
that the endonuclease Hindlll cleaves at
(or near) both ends of the gene, and that
EcoRl cleaves the DNA at one end of
the gene and at a site some distance from
the other end in a flanking sequence. The
fragments made by BamUl resemble
those formed by EcoHl, but the sites are
reversed. The vector pMB9 has one re-
striction site for EcoKl, for Hindlll,
and for BamUl, but the EcoRl site had
been destroyed by the poly (dA) -tail ad-
dition (except in the case of pFb29).
Using this information, we screened the
fibroin gene plasmids, first by testing
whether they had additional restriction
sites for these enzymes and then by
checking whether there were any restric-
tion fragments that did not hybridize
with i25i_jnj^;N'A. This screening method
identified the plasmids carrying both the
mRNA coding and flanking sequences.

Of the 20 plasmids tested, most turned
out to be intragenic fragments of 0.6 to
11 kb. However, pFblO, pFbl9, and
pFb29 revealed multiple restriction sites
for several endonucleases. As shown in
Fig. 46a the pFblO has three Hindlll

DEPARTMENT OF EMBRYOLOGY

69

100 -

E
o

a

X

Temp °C

Fig. 45. The thermal denaturation profiles of pFblO, pMB9, and E. coli DNA. About 5 /ug of
pFblO, pMB9 and E. coli DNA were subjected to the thermal denaturation in 0.1 X SSC.
pFblO and pMB9 DNA were made linear by Hindlll digestion before the denaturation.
— •— , pFblO DNA ; — O— pMB9 DNA ; ,E. coli DNA.

sites, yielding three fragments: A (11.6
kb), B (0.8 kb), and C (0.3 kb). Only
fragment A hybridizes with ^^^I-mRNA
(Fig. 46b). 32p.pMB9 DNA hybridizes
strongly with A and faintly with B.
Therefore, we conclude that all of C and
part of B contain flanking sequences.
The ^2P-pMB9 hybridization experi-
ments also assign the location of the
flanking sequence next to the poly (dA)/
(dT) joining region near the pMB9
Hindlll site. There are no BamUl sites
in pFblO except the one from pMB9
itself (Fig. 47). However, pFblO has
three EcoHl sites in the flanking se-
quence, giving three fragments: A (11.6
kb), B (0.9 kb), and C (0.3 kb) (Fig.
47). Both i25i_j^j^NA and 32p_pMB9
DNA hybridize only with the A frag-
ment. Therefore, the length of the flank-
ing sequence is at least 1.2 kb. From
these cleavage patterns we constructed
a restriction map for pFblO (Fig. 49).
As shown in Fig. 48, the pFbl9 has
four Hindlll sites, giving four fragments :

A (11.8 kb), B (5.7 kb), C (3.4 kb),
and D (3.1 kb). 32p.pMB9 DNA hy-
bridized strongly with the A and weakly
with the C fragments (Fig. 48b). There
was an additional faint hybridization
band between the two fragments. How-
ever, it did not match with B, the ap-
pearance was not reproducible, and the
band was not detectable at lower con-
centrations of 22p.pMB9 DNA. It could
be due to either microheterogeneity of
the pFbl9 or the presence of a contami-
nant. Therefore we located C next to A
in the orientation shown in Fig. 49. ^^^I-
mRNA at 10-20 ng/ml generally hybrid-
izes strongly only with fragment A. When
the mRNA concentration was increased
to 40-50 ng/ml, faint hybridization was
detected in three additional bands (Fig.
48c) ; none of them corresponds to any
major restriction fragment; the one
named B in Fig. 48c was close to the
restriction fragment B (Fig. 42a) but
not identical to it. Furthermore, these
faint bands were not detectable in a

70

CARNEGIE INSTITUTION

Fifr. 46. Tlip agaroso pjol oloctropliorosis and hybridization witli ^~"'I-mRNA of Hindlll frag-
ments of pFblO. (a) Elect rophorosis of (from left, to right) "C-\ DNA fliiidUl dig(>s(, intact,
pFblO, and pFblO Hindlll digest, (b) DNA was transfernnl from the gel shown in (a) to a
Millipore filter and hybridized with ^''I-mRNA.

diffc'iT'Tit batch of ))Fbl9 (Tsujimoto and
Suzuki, unpul)li>li(Ml). If the hybridiza-
tion band B rorrespondod to the HindWl
fragment, the hy})ridiz('d RNA should
represent a rather limited region of the 5'

end of mature mRNA (possibly less than
1 kb out of 17 kb for the entire mRNA
sefjuence) or a presumed precursor to
the inatiu'e mRNA, since pFbl9 was
identified as the plasmid carrying the

DEPARTMENT OF EMBRYOLOGY

71

Fig. 47. The restriction analysis of pFblO by
agarose gel electrophoresis. Samples shown are,
from left to right: (1) intact, (2) HirtdlU
digest, (3) Hindlll and Eco^l digest, (4)
Eco^l and BamWL digest, (5) EcoRl digest,
(6) BamUl digest, and (7) Hnclll digest, of
pFblO. (8) pMB9 Haelll digest, and (9) and
(10) \ DNA Hindlll digest.

flanking sequence to the 5' end of the
gene (as will be described in the next
section). Therefore, the hybridized RNA
was recovered and digested with RNase
Ti, and the resulting oligonucleotides
were fractionated on a DEAE-Sephadex
column. The pattern was indistinguish-
able from the one shown in Fig. 44c. So
we infer that the hybridization corre-
sponds not to any unique sequence of
the 5' end of the fibroin gene transcript
but rather to an internal gene hybridiza-
tion with a smaller Hindlll A fragment
derived possibly from minor heteroge-
neity in the pFbl9 plasmid. The Hindlll
B fragment (5.7 kb) was split into two
pieces by EcoKl digestion (Fig. 48a),

one about 4.1 kb and the other about 1.5
kb. The 1.5-kb fragment and the D frag-
ment (3.1 kb) were also d(.'tected in
Hind\\l-Ec(AU digests of pFb29. By
comparing the restriction maps of pFbl9
and f)Fb29, and considering the distance
between the EcoWl anrl Hindlll sites, it
was possible to [)lace fragment D next
to fragment A (Fig. 49). It was noted
that the Hindlll fragments D, B, and
most of C contain the 5' end flanking
sequence. The restriction maps for pFbl9
and pFb29 were constructerl accordingly
(Fig. 49).

In Fig. 50 we summarize the restriction
map of the fibroin gene constructed
mainly from the cloned plasmids and
partly from the analysis of total genomic
DNA. Figure 50 shows the restriction
maps of only the fibroin gene part in-
serted into the vector.

Determination of Orientation of the
Cloned Fibroin Genes

Hindlll digestion of pFblO and pFbl9
gives a large fragment composed of
nearly equal ])arts mRNA coding se-
quence and pMB9 (fragment A in Fig.
49). It follows that upon digestion of this
fragment with exonuclease III or A-
exonuclease, the coding strand will either
be digested away or be conserved and so
become available for hybridization with
1251-mRNA. We found that digestion of
pFblO with exonuclease III gave a hy-
l)ridizable coding strand without de-
naturation of the digested DNA. There-
fore the 5' end of the coding strand is
conserved and single stranded. We found
also that the coding strand of pFbl9 be-
came hybridizable upon digestion of the
DNA with A-exonuclease, which indicates
that the 3' end of the coding strand is
conserved in this plasmid. Therefore, we
conclude that pFbl9 carries the coding
sequence corresponding to the 5' end of
mRNA and extra sequences adjacent to
the 5' end of the gene. We also conclude
that pFblO contains the sequence coding
for the 3' end of the mRNx\ and an extra
sequence adjacent to it.

72 CARNEGIE INSTITUTION

12 3456789 12 34 56789 123456789

Fig. 48. Restriction analysis of pFbl9 by an agarose gel electrophoresis. Samples shown are
(1) intact, (2) Hiudlll digest, (3) Hindlll and BamHI digest, (4) BamEl digest, (6) EcoUl
and Hindlll digest, (7) EcoRl and BamRl digest, (8) EcoRl digest, (9) Sail digest, and
(5) Sail digest of pFblQ with a size marker of \ DNA Hindlll digest, (a) Electrophoresis pat-
tern; (b) Hybridization of the DNA with 'T-pMB9 DNA; and (c) Hybridization of DNA
derived from a similar experiment with ^^^I-mRNA.

fiomio

^ffMd

Hind III

I EcoRl

AjV ilA/dT(EcoRl)
< Shearing ind

Fig. 40. Restriction maps of the plasmids pFbl9, pFb24, pFblO, and pFb29. Numbers in the
figure stand for kilobases.

DEPARTMENT OF EMBRYOLOGY

73

Gene

^3

! 1
1 I

17 kb

1^

K—

1 •

k

'*'

21 kb

k H^

2.9

H+K-*

i >K

4.1 1.5 3.1

pFb19

pFblO

6.0

.5.8

pFb24

pFb29

21

Fig. 50. Restriction map of fibroin gene compiled from plasmids and B. mori total DXA,
together with restriction maps of fibroin genes in the plasmids. i Hindlll site, | EcoRl
site, ; BamEl site, | sheared DNA ends with dA/dT.

Identification of the Site and

Sequence of the Primary Transcript

OF THE Fibroin Gene

Y. Suzuki, R. Reeder, Y . Tsujimoto,
and Y . Ohshima

Before we proceed to the sequence
analysis of the presumed regulatory re-
gion and faithful transcription studies of
the gene in vitro we have to locate the
site of primary transcription on the gene
map.

Hybridization of i^si.^j^;^^ ^^ 20
ng/ml in l-/>tg samples of the plasmids
pFbl9 and pFb29 occurs only with the
largest Hindlll fragments (A). From
the restriction mapping described above
and from the hybridization experiments,
we conclude that the 5' terminus of ma-

ture fibroin mRNA is very close to the
Hindlll site of the A fragment (within
1 kb). Our i25i_j^j^]v^T^ preparation in-
cludes putative precursor molecules of
the mature mRNA because in the mRNA
purification procedure we always pool
whole RNA fractions greater than 40S.
With the use of this mRNA preparation,
other Hindlll fragments (B. C, and D)
did not show any hybridization even
after a prolonged incubation at an
mRNA concentration of 50 ng/ml. How-
ever, we cannot locate the site of the pre-
sumed primary transcript with this ex-
periment because we do not know the
concentration of putative precursors in
the mRNA preparation.

Yang et al, in Cell 7, 339 (1976),
reported that only 60% of fibroin mRNA

74

CARNEGIE INSTITUTION

is found capped. But a group oi enzymes
from vaccinia is known to cap RNA ends
having diphosphate or triphosphate
(!Moss. Biochoti. Biophijs. Res. Comm.
74, 374. 1977). A fibroin mRXA prepa-
ration containing putative precursors was
capped by these enzymes and hibeled
with S-adenosyl-(^H) methionine in vitro.
The hibeled mRXA was hybridized to
Hindlll fragments from three kinds of
fibroin gene phismids: (1) pFbl9 having
12 kb of the 5' end flanking sequence, (2)
]iFb2o having only the intragenic por-
tion, and (3) pFblO having 1.3 kb of the
3' end flanking sequence. The hybrids
were exposed to RXase Tx to eUminate
nonhybridized mRXA. As shown in Fig.
51. the '^H-mRXA livbridized only with

the Hindlll A fragment of pFbl9. The
specificity of this hybridization confirms
that methylation occurred at the 5' end
of niRXW (or precursors). It suggests
also that the primary site of transcription
is within the Hindlll A fragment and
that any niRX^A precursor, if one exists,
can not be longer at its 5' end than
mature mRXA within the limit of ex-
perimental error (less than 1 kb, which
is about 6% of mature mRXA length).
The RXA hybridized to the Hindlll A
fragment was recovered and digested
with RXase Ti, and the capped structure
was analyzed on a DEAE-Sephadex col-
umn. There is only one component having
about a — 8 charge, which is quite simi-
lar to the charge of the major capped

200

150

100

50

50

50

• 1 11

A B CD

If* it'^0_^^te

:ft^

• • ■^

"*■•••■'•

-fil^

8

c m

12

16

Fi^. 51. Hybridization of Hindlll fragments of pFblO, pFb25, and pFblO with ^H-mRNA
capped in vitro by the vaccinia enzymes. 10 Mg each of pFblQ, pFb25, and pFhlO were digested
with Hindlll, elect^opho^esed, transferred to a Millipore filter, and hybridized with ^H-mRNA
which had been capped in vitro by the vaccinia enzymes (a generous gift from B. Moss). The
electrophoresis pattern was shown schematically above the radioactivity profile in each panel.
C — • — ), cts/min.

DEPARTMENT OF EMBRYOLOGY 75

structure of the input mRNA and that of terminal portion of the fibroin DNA

in vivo capped mRNA. because this region is likely to be in-

In order to determine whether or not volved in the control of transcription,

the initiation of chromatin transcription Nucleotide sequence analysis of DNA

in vitro is faithful, we will compare the requires a pure preparation of specific

5' end sequence of the mature mRNA DNA fragments. Fibroin DNA with 5'

and the in vitro capped mRNA, localize flanking has provided a source for large

the site of transcription origin, and se- amounts of pure DNA fragments that

quence the surrounding region. are amenable to sequencing.

A detailed restriction enzyme map

near the 5' end of the Hindlll site of

Sequencing of a Presumed Regulatory pp^^g ^^n ^^ constructed, and appro-

Region of the Fibroin Gene priate restriction fragments sequenced.

Y.Tsujimoto,Y.Ohshim.a, and Y.Suzuki ^e also plan sequencing analysis into

the supposed spacer region as well as the
We are interested in the nucleotide 3' end of the gene and its flanking
sequence of the region adjacent to the 5' sequence.

RIBOSOMAL DNA AND RELATED SEQUENCES IN

Drosophila melanogaster

I. B. Dawid

In Year Book 75, pp. 30-36, we re- Studies of Cloned rDNA Repeats
ported the isolation and characterization

of ribosomal DNA (rDNA) from D. I.B.Daimdincollahorationwith

I i\. W bIlcluct

melanogaster. This DNA is composed of ^-^^ ^j^^ assistance of M. Rebhert

repeating units that occur in two basic

structures. One is similar to the rDNA Purified or enriched rDNA was di-

unit of Xenopus: It has a region that gested with the restriction endonuclease

codes for the ribosomal RNA precursor ^coRI, and the fragments were attached

and a spacer that separates this region to the vector pMB9. The recombinant

from the next coding segment. About DNA was introduced into E. coli, and

two thirds of the repeats differ from the clones containing the ribosomal DNA

first class. A DNA sequence which does were selected by techniques similar to

not code for rDNA is inserted in the 2SS those described in Year Book 74, p. 20.

genes of these repeats. The functional Most rDNA repeats have a single EcoRl

implications of this insertion sequence site, and therefore most EcoHl fragments

are not known. We have shown that in- of rDNA are full repeating units. We

sertion sequences occur in different size obtained a number of clones containing

classes with a predominant size of 5 full rDNA repeats. In addition, there

kilobases (kb). In the past year, we have are repeats of rDNA that have a second

analyzed the structure of rDNA repeat- ^coRI site within the insertion in the

ing units in more detail, using recom- 28»S RNA gene. Consequently, some frag-

binant DNA technology, and we have ments produced by Eco^l are '4ialf-

studied sequences homologous to the repeats." We have isolated four recom-

ribosomal insertion in other parts of the binant plasmid molecules that contain
Drosophila genome. The results of these

studies show that the ribosomal locus in ^g^^.^^^ 1^^^^^^^^^ ^^^ Experimental Cancer

Drosophila is exceedingly complex. Research, Lausanne.

76

CARNEGIE INSTITUTION

TABLE 7. Summan' of Properties of Cloned rDXA Fragments from Drosophila melanogaster

Class

Number oi
Clones

Size of rDXA
Fragments*

Insertion
Size

Restriction Sites
in Insertion

I
II

in

IV
V

2
5
9
4

lo.o-l"
II -12
11 -12
11 -12
4.3- 7.4

5 Smal{2),Bam'Rl{2)

1.0 BamBl (2)

0.5 Bamm (1)

None
Eco^l site in insertion giving
lialf-repeats. No Sma or Bayn site.

* All sizes are given in kilobase pairs (kb).

approximatoly half of a repeating unit
of rDXA (Table 7 and Fig. 52).

To characterize these repeating units,
we carried out some restriction mapping
and a series of hybridization experi-
ments with cloned rDNA's. A restriction
map for one rDXA repeat containing a
long insertion is shown in Fig. 53. Several
other clones with short insertions or no
insertions were also mapped. Table 7
summarizes the major features of these

other clones. Some clones lacked any
insertion in the 28*S gene. These clones
have no sites for the endonuclease
/)a//?HI. They also lack those sites for
Smal which are present in the insertion
(Fig. 53). The sites in the other parts
of the molecule, the transcribed regions
and the spacer region, are at analogous
positions in rei)eats with and without in-
sertions. Several clones contained inser-
tions of 0.5 kb, as determined by electron

Fig. 52. Gel f'leftrophorosis of rlonrd Drosophila rDNA after digestion with EcoTlJ. The first
lane in r-ach panel contains marker fragments obtained by digestion with nirullll of i)hag(^
X DNA. The othor lanes are recombinant elonos, racli containing the vector pMB9 (arrow)
and a fragment of rDXA. Some recombinant clones contain sev(>ral fragments of DNA. Tlu>
positions of the major EcoKl fragments of Drosophila rDNA are indicated at 11 and 16 kb.

DEPARTMENT OF EMBRYOLOGY

77

18S

15 KB

Fig. 53. Map of the cloned rDNA fragment
Dmra56. The ends of the niole(;ule coric^spond
to EcoHl sites that were used in cloning tlK>
fragment. Smal (arrows) and BamHl (arrow-
heads) sites are shown below, and Ifitidlll
(arrows) and Haclll (arrowheads) sites are
shown above. All Smal, BamHI, and Hindlll
sites are shown but the Haelll ma]) is incom-
plete. There are no additional Hnelll sites be-
tween the two Haelll sites shown so that a
large fragment containing the spacer may be
obtained. There are many Haelll sites else-
where in the DNA which have not been
mapped. The Smal D fragment extends from
4.3 to 6.6 kb; the Hindlll B fragment is from
4 to 9 kb.

microscopy. These insertions also con-
tained a single BamUl site in the in-
sertion. For further characterization of
the .relations of different insertions we
used restriction fragments derived from
the clone Dmra56 (Fig. 53) and hy-
bridized these fragments with the other
rDNA clones. In particular, we used the
Sma D fragment and the BamUl frag-
ment. Clones with 0.5-kb insertions and
a single Bam site hybridized well with
the BainUl fragment of Dmra56 but did
not hybridize with the Sma D fragment,
nor did they hybridize with the long
insertion fragment that is derived from a
Hindlll-Bam double digest and goes
from 4.0 to 8.0 kb on the map of Fig. 53.
Therefore, we conclude that clones with
0.5-kb insertions contain insertion se-
quences homologous to the rightmost
500 base pairs of Dmra56. Two other
clones contained 1.0-kb insertions. These
clones had two BamUl sites and hy-
bridized with the BamUl fragment of
Dmra56 but not with the Sma D frag-
ment. The Sma D fragment is not homol-
ogous to the BamUl insertion fragment
of Dmra56. In addition to Dmra56, we
obtained two other clones with long in-
sertions (5 kb). These clones are entirely
homologous in structure to Dmra56.

We did not obtain any recombinant
clones with insertions of intermediate
length (2-4 kb). The reason for this can
I)robably be found in fragments which
contain only about half of a ref)eating
unit. Four such clones were obtained
(Table 7). From an analysis of rDNA
from the Y chromosome (see l)elow) we
infer that most insertion sequence's of
intermediate size contain an Ecolll site.
Such insertion repeats are likely to have
given rise to the four clones listed at the
bottom of Ta})lc 7. The most intriguing
aspect of the structure of these clones
with incomplete repeats is the apparent
absence of homology with any part of
the insertion sequence of Dmra56. All
four "half-repeat" rDNA clones were
hybridized with labeled fragments de-
rived from all parts of the insertion of
Dmra56, and no significant hybridization
was observed. We have not yet carried
out this hybridization at different cri-
teria, but it is clear that homology
between the 5-kb insertion and the inter-
mediate-size EcoTll restrictable insertions
is either low or absent. In contrast the
short insertions of 0.5 and 1 kb are ho-
mologous to the right-hand side of the
5-kb insertions. These experiments sug-
gest that there are two different classes
of insertions in Drosophila rDNA: in-
sertions with and insertions without an
EcoRl site. These two classes appear to
})e nonhomologous. The class without an
Ecolll site, and probably the other class
as well, occurs in different lengths.

RiBosoMAL DNA Insertions in the

Y Chromosome Nucleolus Organizer

OF Drosophila

I. B. Dawid in collaboration with K. D. Tartof*
andP.K. Wellauerf

We reported earlier (Year Book 75,
p. 36) that rDNA repeats with the char-
acteristic 5-kb insertion are present in

* Cancer Research Institute. Philadelphia.
t Swiss Institute for Experimental Cancer Re-
search, Lausanne.

78

CARNEGIE INSTITUTION

the nucleolus organizer on the X chromo-
some but absent or very rare on the Y
chromosome. These observations were
based on gel electrophoresis of DNA
digested with EroRI. In the EcoTil prod-
ucts derived from the Y chromosome, we
did observe smaller bands of rDNA
which were shorter than a single repeat-
ing unit. In view of the presence of
f'roRI sites in some insertions it seems
possible that these smaller fragments
seen in rDNA from the Y chromosome
could be due to repeats with intermediate-
size insertions. This is, in fact, the case.

DXA was isolated from flies that carry
the scute 4 scute 8 inversion on the X
chromosome and thus derive all their
rDXA from the Y chromosome. The
rDXA was partially purified and then
analyzed by electron microscopy after
hybridization with rDXA. In this experi-
ment. 169c of all repeating units were
found to contain an insertion in the 28*S
gene. This insertion was at the same
location as the insertions in total rDNA
from wild-type flies analyzed earlier.
However, the size distribution of these
insertions was different from those in
total Drosophila DNA: Most insertions
had an intermediate size between 2 and
4 kb, and few 5-kb insertions were found.
This result means that rDXA on the Y
chromosome does carry insertions but
that these insertions have a different size
distribution than the insertions on the X
chromosome. Furthermore, the absence
of EroVvl fragments longer than the con-
tinuous repeating unit of about 11 kb
from Y rDXA is explained if we assume
that all or most of the insertions on the
Y chromosome have an extra EcoTil site.
This assumption is in good agreement
with the results of the cloning experiment
reported above. In fact, two of the four
clones with "half-repeats" were derived
from rDN^A from the Y chromosome.

We also studied the sequence homology
of the spacer region in rDNA derived
from the X or Y chromosome. A spacer
region, distinct from the insertion, is
present in every repeat of rDNA. Re-

sults from both heteroduplex molecules
analyzed by electron microscopy and
hybridization experiments with radio-
actively labeled spacer sequences sug-
gest that the spacers on X and Y rDNA
are homologous and form stable duplexes.
The precise degree of homology has not
yet been investigated in sufficient detail
to allow us to say whether or not signifi-
cant mismatching occurs. It is clear,
however, that rDNA spacers on the two
chromosomes are closely related.

In the experiments described so far,
the ribosomal locus in Drosophila is
viewed as a highly complex set of se-
quences. While the transcribed region
appears to be conserved in all repeating
units, the spacers vary in length and,
in addition, a large fraction of the genes
for 28*S RNA have an insertion. Inser-
tions occur in two distinct sequence
types: One is limited to the X chromo-
some, while the other appears to occur
on both the X and Y chromosome. Both
sequence classes of insertions occur in
different size classes. While it is possible
that internally repetitious regions occur
within the insertion it is clear that a 5-kb
insertion is not composed entirely of re-
peated sequences. The evolutionary rea-
sons for this complexity, and its func-
tional implications, are not clear at
present.

Sequences Homologous to Ribosomal

Insertions Occur Elsewhere in the

Drosophila Genome

7. B. Dawid and P. Botchan
with the assistance of M. Rebbert

We wondered how insertion sequences
came to be present within the ribosomal
genes. If at some point sequences that
were present elsewhere did become in-
serted into the ribosomal genes, it would
be conceivable that similar sequences
were also inserted into other locations of
the Drosophila genome. Therefore we
tested for the presence of homologous
sequences at extraribosomal sites. For

DEPARTMENT OF EMBRYOLOGY

79

this purpose wc used restriction frag-
ments composed entirely of rihosomal
insertion sequences and derived from the
cloned rDNA fragment Dmra56 (Fig.
53) . The Sma D fragment and the BamHI
fragment, which represent different non-
homologous parts of the insertion, were
used as probes after being labeled with
^^P by the nick translation method.

We found that non-rDNA insertion
DNA does occur in the Drosophila
genome and that this DNA could be
separated from rDNA in several types
of density gradients. In Fig. 54 total
DNA from Drosophila embryos was
banded in a cesium chloride gradient
containing actinomycin D. Aliquots of
each fraction were hybridized with rRNA
to locate the rDNA, and with ^^p-labeled
Smal D fragment or ^^F-BamUl frag-

E

Q.

b

a

Q.

U

t— t

„

CM

CL

CJ

to

10 20

Fraction Number

Fig. 54. Total DNA from Drosophila em-
bryos was centrifuged to equilibrium in a gra-
dient of CsCl containing actinomycin D. A
small amount of ^T-labeled Smal D fragment
from pDmra56 (Fig. 53) was added to the
gradient at the outset. Its position is shown by
closed triangles (right-hand scale), and marks
the density of pure insertion sequences. Ultra-
violet absorbance, marking the position of
main-band DNA, is shown as a thin Hne. Small
samples of each fraction were hybridized with
^^-rRNA (•-•), ''F-Smal D fragment
(O-O), or 'T-BamHI fragment (A-A) of
pDmra56. Hybridization with these fragments
locates insertion sequences in the gradient.

ment of Dmra56 (compare Fig. 53j. The
rDNA bands on the dense side of the
main band and hybridizes with the Sma
and Bam fragments. Other DNA com-
ponents that are homologous: to insertion
sequences band on the low density side
of the gradient.

In this experiment, we added at the
outset a small amount of labeled Sw.a D
fragment of Dmra56 to mark the position
of pure insertion sequences in the gradi-
ent. The pure Sma D banded at low den-
sity, even lower than most of the non-
rDNA components that hybridized with
the insertion probes. The best interpre-
tation of this result, supported by addi-
tional data not discussed here, is that
blocks of insertion sequences are linked
to DNA of different base composition in
the non-rDNA insertion DNA compon-
ents. Pure insertion has a nucleotide
composition quite different from that of
the main-band DNA of Drosophila; its
GC content is close to 55% while main-
band DNA has 40% GC.

Figure 55 shows a similar gradient in
CsCl containing the dye Hoechst 33258.
Here the rDNA is on the low density
side of the gradient, again separated
from DNA components that hybridize
with the insertion probes. Pure insertion
sequences, here marked with the Bam
fragment of pDmra56, band at a higher
density than most non-rDNA insertion
DNA components from Drosophila.

Gradients like those shown in Figs. 54
and 55 could be used to separate non-
rDNA insertion DNA from rDNA and
to enrich it with respect to main band.
Such purified DNA was analyzed with
several restriction endonucleases. For
this purpose, the DNA was digested,
separated on agarose gels, and trans-
ferred to membrane filters by Southern's
method. Fragments were then hybridized
with different probes. Figure 56 shows
the results of such an experiment after
digestion of the DNA with BainHl.
When total embryo DNA was digested
to locate the rDNA, we obtained the
picture shown in lane a. Since only the

so

CARNEGIE INSTITUTION

2 -

lO

E

Q.
U

Q.
CM
lO

Bam

1

rRNA
MB.,,

■

i fl

c^

' P'

- 8

- 6

- 4

- 2

E
u

CM

ro

10 20 30

Fraction Number

Fig. 55. Drosophilu embryo DNA was banded
in CsCl containing the dye Hoechst 33258.
Labeled BamUl fragment from pDmra56 was
added as a density marker (closed triangles
righthand scale). The positions of main-band
DXA (M.B.) and of rDNA (visuahzed by
hybridization with rRNA) are indicated by
arrows. Aliquots of the fractions were hy-
bridized with ^'F-Smal D fragment (open
circles), or ^'F-BamHl fragment (open tri-
angles) .

insertion has a Bam site, most rDNA
molecules are not cut into distinct frag-
ments but give rise to a smear of high
molecular ^veight products. Hybridiza-
tion of the same digest with insertion
probes (lanes b,h,i) resulted not only in
this smear produced by the rDNA but
also in a number of sharp fragments.
With purified non-rDNA insertion DNA
we obtain only distinct, sharp bands
f lanes c-g, j-l). Many fragments are
produced in different sizes (^20-1 kb).
The fragments hybridize with insertion
probes to very different degrees, which
suggests that some occur rarely, perhaps
only once per genome, while others occur
manv times.

Digestion of non-rDNA insertion DNA
with three other restriction enzymes,
£^coRI, Hindlll, and Stnal, and hybridi-
zation of the resulting fragments with
the same insertion probes led to basically
comparable patterns. While the actual
distribution of fragments was different
in all cases, many different fragments
occurred, and there was no resemblance
to the pattern produced from rDNA by
the same restriction endonucleases.

The result shown in Fig. 56 and also
obtained with the other restriction endo-
nucleases demonstrated that very similar
fragments were obtained from non-rDNA
insertion DNA derived from embryos,
pupae, or adult Drosophila. Furthermore,
three different methods of purification
led to the same restriction patterns. The
patterns, as visualized by the Sma D
or the BamUl insertion probes, were
quite similar in the high molecular
weight range but differed among smaller
fragments.

We interpret these results to mean that
sequences closely related to the ribo-
somal insertion appear in other regions
of the Drosophila genome. There are
about equal amounts of non-rDNA and
rDNA insertion sequences in the genome.
The amount outside the ribosomal locus
corresponds to about 400 kb of sequence,
or 0.2% of the genome. Several indica-
tions suggest that non-rDNA insertion
sequences are in blocks that may have
sizes similar to the ribosomal insertions:
from perhaps as little as a few hundred
base pairs up to 5 and possibly as long
as 10 kb. These insertion sequences are
linked to DNA of different base composi-
tion; its nature is unknown except that
it is very likely that these insertion-
linked DNA components are diverse
types of sequences. Non-rDNA insertion
DNA occurs in different stages of the
Drosophila life cycle, and there is no
major change and perhaps no change at
all in the abundance and distribution of
this DNA in the different stages.

DEPARTMENT OF EMBRYOLOGY

81

abcdefgh ij k I

o-

0.9

Fig. 56. Restriction patterns of Drosophila DNA after digestion with BamJU.. Lanes a-h
were run on one gel slab, i-l on another; the proper line-up is indicated between lanes h and
i. Hybridization was with ^^^I-rRNA (a), ^^P-BamHI fragment (b-g), and ^-F-Smal fragment
(h-l). We used total DNA from embryos (a, h, i) or adults (b) ; or non-rDNA insertion DXA
from embryos (c, d, j, k), from pupae (e, /), or from adults {g, I). Purification was done in
gradients containing Hoechst (c, e, k), actinomycin D (/, g, I), or mercury (d, j).

82

CARNEGIE INSTITUTION

BIOGENESIS OF MITOCHONDRIA

Igor B. Dawid, E. A. Godicin. and J . L. Ramirez
u-ith the technical assistance of M . L. Rebberl

Our interest in this area has been
focused on the mapping of the mitochon-
drial genome of Xenopus laevis. The prod-
ucts of mitochondrial DNA (mtDNA)
in animal cells include two specific
rRNA's and a set of tRNA's. The sites
on the heavy (H) strand that code for
these gene products have been mapped
previously {Year Book 75, pp. 45-49).
In the past year, we mapped the sites
that code for mitochondrial 4S RNA on
the light (L) strand of mtDNA. We
initiated experiments to characterize
l)oly(Ai containing mitochondrial RNA

molecules, from ovaries of X. laevis. Such
molecules occur in mitochondria and are
probably messenger RNA's for mito-
chondrially synthesized proteins.

]\L\ppiNG OF 4<S RNA Genes in
X. Laevis mtDNA

./, L. Ramirez

In Year Book 75, pp. 48 and 49, we
described the mapping of the 4»S RNA
genes on the H strand of X. laevis
mtDNA. Sites on the L strand have now
been mapped, using an EcoKi fragment

Fig. 57. Eloctron micrograph of tlio L strand of tho largo EcoKi fragment of X . lacim
mtDNA carrying three complexes })etween 4S RNA and ferritin. A fragment ])ro(hi('ecl from
mtDNA by digestion with II pall was used as a marker of orientation; hybridization with this
marker produced the duplex region at the left of the molcHuiie as indicated by a heavy line
on the tracing. The ferritin closest to the marker is at position LI ; the other two ferritins are
at po.sitions considered tentative (see text and Fig. 59).

DEPARTMENT OF EMBRYOLOGY

83

40
Percent of

60
Genome

EcoR l

100

Fig. 58. Histogram of ferritin^S RNA sites on the L strand of X. laevis mtDNA. The map
begins at the EcoHl site closest to the ribosomal gene region (see Year Book 75, p. 47). The
six sites are indicated by large arrows; four sites considered tentative are shown by small
arrows. The duplex formed by the Hpall fragment used as a marker is shown as an open
region in the histogram.

that represents 87% of the mitochondrial
genome (see Year Book 75, p. 47). This
DNA fragment was cloned in E. coli
cells. Because the ribonucleotides in
mtDNA are eliminated by the cloning
procedure, the DNA is stable in alkali,
and so it is possible to separate its
strands in alkaline CsCl gradients

The 45 RNA's from X. laevis ovarian
mitochondria were labeled with ferritin
as described in Year Book 75; the con-
ditions for hybridization were also the
same. Excess ferritin was removed by
chromatography on a small hydroxylapa-
tite column equilibrated in 0.15 M Na
PO4 buffer, pH 7. Under these conditions,
ferritin elutes in the void volume, while
the DNA-45 RNA hybrids are retained
and eluted later with a solution of 0.5
M NaCl, 0.5 M NaP04, pH 7.

Electron-microscope analysis of the
hybrid molecules (Fig. 57) gave the
histogram shown in Fig. 58. Six sites have
been unambiguously defined on the basis
of their position relative to the marker
(Hpall fragment). Four other sites are
indicated that probably code for 4*S RNA.

Figure 59 summarizes the results of
the 45 RNA mapping experiment. The
two strands were aligned using markers

(see Fig. 58, legend). The H strand map
is from Year Book 75, p. 49. The total
number of 4*S RNA sites on mtDNA is
between 21 and 25, close to the number

H9 H!

Fig. 59. Map of iS RNA sites in Xenopus
laevis mtDNA. The ribosomal genes on the H
strand are shown as heavy bars; the antiribo-
somal regions on the L strand are sho'WTi as
open bars. Closed circles are positively identi-
fied 4^ RNA sites; open circles are tentative
sites. The H strand map is taken from Year
Book 75, p. 49.

S4 CARNEGIE INSTITUTION

required for an independent set of mito- by Angerer et al. {Cell 9, 81, 1976).
chondrial tRNA's. There are no long regions of DNA with-
The nearly uniform distribution of 4*S out a 4S RNA site on one strand or the
RNA sites over the mtDXA is empha- other. It remains to be seen whether the
sized even more strongly when both other products of mtDNA (presumed
strands are considered. This wide distri- mRNA's) can be fitted between the 4^S
bution of 4S RNA sites is similar to the RNA genes or whether overlapping cod-
distribution found for HeLa cell mtDNA ing regions occur in this DNA.

RIBOSOMAL GENES AND THEIR PROTEINS

Ronald H. Reeder, Peter Botchan, Harvey Wahn,

Robert Hipskiiid, and Barbara Sollner-Webb

with the assistance of Eileen Hogan

For several years our group has fo- structure, they are clearly distinct from
cused its attention on the genes that code each other. These studies have also shown
for the 18S and 28S rRNA's in the frog that there is a sharp boundary close to
Xenopus. Our immediate goal is to pro- the 5' end of the 405 sequence. Se-
vide a complete molecular description of quences within the transcribed region are
these genes, both of the DNA sequence strongly conserved, while sequences out-
itself and of the proteins that interact side (in the spacer) vary even within a
with it in vivo. Determination of the species. If it is reasonable to assume
sequence organization of rDNA has been that promoter sequences are conserved,
greatly aided by recent advances in DNA then this conserved-nonconserved bound-
technology. With the methods of re- ary should help define the location of
striction enzymology, molecular cloning, the promotor.

and rapid nucleotide sequencing now As another approach to locating pro-
available, it is becoming routine to estab- moter sequences, we used an enzyme
lish the primary sequence of DNA mole- complex from Vaccinia virus which puts
cules. It is still a major challenge, how- radioactive ''caps" on 5' diphosphate or
ever, to determine where the various con- triphosphate termini of RNA molecules,
trol elements are located within a DNA On the assumption that such termini are
sequence, especially when a single gene the residue of chain initiation starts by
plus spacer contains over 10,000 base RNA polymerase, we have capped nucleo-
pairs, as in Xenopus rDNA. Rather than lar RNA and mapped regions in the
blindly sequencing this large repeat, we rDNA that code for the capped termini,
have spent considerable effort in pre- The results so far indicate that 85% of
liminary mapping to determine where all chain initiations on rDNA begin at
important control sequences are most or near the 5' end of the 40>S sequence,
likely to be found. The capping procedure has also allowed

Part of this work continues the re- us to begin sequencing the 5' end of the

striction mapping of the nontranscribed 40*S molecule for later comparison with

spacer that was begun last year. In cor- the underlying DNA sequence,

roboration of earlier heteroduplex studies, A third approach, in collaboration with

the restriction work has accurately de- Dr. Robert Benbow at Johns Hopkins,

fined the boundaries of the two regions has been used to search for possible ori-

of simple-sequence repetitive elements in gins of DNA replication in the rDNA

the spacer and shown in addition that repeat. Chimeric plasmids containing

while the two regions are related in various regions of the rDNA repeat have

DEPARTMENT OF EMBRYOLOGY 85

been observed to replicate in an in vitro peated sequences and thus resembles
system developed by Dr. Benbow, but simple-sequence satellite DNA's. Super-
it is too early to say whether replication imposed on short repeating sequences are
is initiating at the correct site. All of levels of higher order repeating elements,
these studies have paved the way for Restriction endonucleases that have
sequencing selected portions of the rDNA, many sites in the spacers (such as HaeWl
which will be done during the coming and Hpall) digest most parts of the
year. spacer of all four clones into a similar
Determination of the proteins that in- set of repetitive fragments. Most of these
teract with the rDNA is as important as fragments have sizes that are multiples
determination of the primary sequence, of about 15 bp; this leads us to believe
It is now possible to catalogue the rDNA- that the spacer comes from an ancestral
associated proteins, using the method 15-bp unit.

reported last year for isolating amplified Figure 60 summarizes our present view
oocyte nucleoli that contain rDNA but of the structure of a typical spacer. Re-
no bulk chromosomal DNA. Efforts this striction enzymes that recognize few
year have been directed toward improv- sites (such as BamHl and Smal) allow
ing the isolation procedure and over- the conclusion that the spacers are sub-
coming the effect of an endogenous pro- divided into three distinguishable regions,
tease that contaminates nucleoli. Region A is immediately adjacent to the
Antibodies directed against specific 28^ end of the gene near the transcrip-
chromatin proteins promise to become tion termination site, is 500 bp long, and
useful tools for studying the distribution is conserved among the four clones exam-
of various proteins on ribosomal gene ined. Adjacent to it is Region B, which
chromatin. As a step in developing these has multiple sites for Smal. Region B
tools we have collaborated with Dr. varies in length from 1(X)0 to 1200 bp
Michael Bustin (NIH) and Mr. Steven among the four clones. The remnant,
McKnight (graduate student with Dr. region D, has few Stnal sites and varies
Oscar Miller, Jr., University of Virginia) from 890 to 3800 bp in length. In long
in comparing the immunological related- spacers, region D has a higher order re-
ness of calf thymus and Drosophila his- peating unit (as recognized by BamUl
tones. In the following pages, each of and Smal). Although regions B and D
these projects is described in more detail, may have arisen from the same ancestral

15-bp repeat, the sequences in these re-
gions have since diverged because of an

Structure OF THE NoNTRANSCRiBED apparent barrier at the B-D boundarv.

Spacers between Ribosomal Genes ^e are trying to find out whether the

Peter Botchan ribosomal spacer shares any homology

with nonribosomal DNA. In particular.

We have continued our study of se- we want to know whether the ribosomal

quence organization in the nontranscribed spacer, which is itself a middle-repetitive

spacer regions of four cloned fragments sequence, is related to some family of

of Xenopus laevis ribosomal DNA. Each highly reiterated sequences. Or, are

of the clones we are using contains a spacer-like sequences located anywhere

spacer of different length, and we have at all outside of the ribosomal gene

mapped the distribution of cleavage sites cluster. This question has some bearing

for several restriction endonucleases in on the theories which suggest that middle-

these spacers. Some of our findings and repetitive sequences have some role in

techniques were reported in Year Book coordinate gene expression. To search for

75 (p. 20). We now conclude that much such homologies, we fractionated red

of each spacer is composed of highly re- blood cell DNA derived from single frogs

86

CARNEGIE INSTITUTION

Xlrl4

RI Hindin

RI

28S A B C D

«<— Tsp + I8S —

►

500^^ 900 800 3380

2530

2120 750 660 660 660

3400

1000

Conserved Region

Fig. 60. Summan' of structural data for a cloned spacer from X. laevis rDNA. These maps
summarize data from previous heteroduplex mapping (Year Book 73) and from BarnRl and
Smal restriction analysis reported this year. The EM heteroduplex map has been redrawn with
the following modifications: (1) The amount of 285 on the left has been reduced to 500 bp. The
Hindlll site is within 50 bp of the transcription termination site (details not shown). (2)
Region X in Xlrl4 (a region of about 350 bp just to the left of the 5' end of the gene) has
been included in region D. All known BamHl and Smal sites are represented, except the Smal
sites within the 1000-bp region that coincides with region B. Sites in this region are too finely
spaced to draw accurately. Molecular weights are in bp.

by equilibrium density centrifugation in
CsCl. Each fraction from the gradient
was then cleaved with EcoRl endo-
nuclease and electrophoresed on an
agarose gel, and the separated fragments
were transferred to a nitrocellulose filter,
using Southern's procedure. The filter-
immobilized DXA was then hybridized
with a highly radioactive probe DNA to
locate rDXA sequences. The probe we
used was a cloned EcoRl spacer frag-
ment that contains all of the non-
transcribed spacer plus some gene se-
quences on each end. It lacks the
sequences contained in the EcoRl frag-
ment from the middle of the gene. This
spacer-containing fragment was labeled
to high specific activity (10^-5 X 10'''
cts/min//Ag) by nick translation and is
able to detect sequences present at 1
copy per genome. The results of a typical
experiment are shown in Fig. 61. Most
of the rDNA fragments that hybridize
with the radioactive probe are associated
with the peak of rDNA on the high-
density side of the gradient. We conclude
that these fragments are components of
the main rDNA locus. Three other frag-

ments can be seen (A, B, and C) which
behave as though they are linked to
DNA of a density lower than that of
rDNA. Fragments B and C are present
about once per genome and band sharply
on the heavy side of the main-band
DNA. Fragment A is present about five
times per genome and bands broadly
between the rDNA and main-band DNA.
Other experiments have shown that frag-
ments A, B, and C do not hybridize with
pure spacer sequences. Therefore, the
hybridization seen in Fig. 61 is presum-
ably due to the 18aS and 28*S gene se-
quences on the ends of the EcoRl spacer
fragment used as a probe. We have not
yet found any rDNA spacer sequences
outside the main rDNA locus. We tenta-
tively estimate that if such DNA does
exist, it would have to be present in
contiguous lengths of less than 500 bp.
This contrasts with the situation seen in
Drosophila rDNA (see Dawid's report,
this Year Book). Fragments of the same
size and density as A, B, and C have
been observed in four different frogs. In
view of the considerable variation in
length seen in the rest of the rDNA re-

DEPARTMENT OF EMBRYOLOGY

87

2

Fig. 61. EcoRl restriction analysis of buoyant density fractionated rDNA from X. laevis.
Bulk DNA from erythrocytes of X. laevis was banded in a CsCl density gradient, and DXA
from each fraction was digested with EcoHl endonuclease. Each digest was electrophoresed on
a 1% agarose gel, and the DNA fragments were transferred to a nitrocellulose filter using
Southern's procedure. The filter was then hybridized with a radioactive probe made by nick-
translating a cloned EcoKL spacer containing fragment (Xlrl2, sp. act. -—'10' cts/min//ig).
Arrows indicate the positions of rDNA and main-band DNA in the gradient, a, h, and c are
hybridizing fragments whose buoyant density position does not correspond to that of the
main rDNA locus.

peats, we speculate that these conserved
rDNA sequences may have some unique
developmental or evolutionary role.

In vivo Sites of RNA Chain
Initiation on rDNA

Ronald H. Reeder, Barbara Sollner-W ebb ,
and Harvey Wahn

The size of the primary transcription
product is still unknown for most eukary-
otic genes that have been studied. Only
in the case of the 5S RNA from ribo-
somes has it been possible to demonstrate
a triphosphate terminus on the 5' end
of the RNA and therefore to pinpoint
the site of chain initiation. In all other
cases, the approach has been to isolate

the largest RNA that can be detected
after brief labeling and assume that is
the primary transcript. Experience with
prokaryotes has shown, however, that
processing may occur so rapidly for
some genes that under normal circum-
stances the primary transcript never
appears intact and the earliest discernible
products may already be processed in-
termediates.

When cells of Xenopus laevis are
labeled with RNA precursors, the first
distinct transcription product that arises
from the rDNA is a molecule of 2.7 X
10^, the so-called 40S rRNA precursor.
Through a series of known processing
steps this precursor then gives rise to
the 18S and 28iS rRNA's found in cyto-

88

CARNEGIE INSTITUTION

plasmic ribosomes (see Year Book 73,
p. 45V The assumption has been, there-
fore, that transcription initiates close to
the 5' end of the 40.S sequence. However,
several experiments from other labora-
tories have raised the possibility that
transcription of the rDXA does not begin
at the 5' end of the 405 sequence but
somewhere else in the ''non-transcribed"
spacer that separates 40-S regions.

We have been using a new method to
search for and characterize transcription
initiation events on the rDNA of X.
laevis. The method involves use of a
complex of capping enzymes from Vac-
cinia virus which catalyze the reaction
pppG+S-adenosvl methionine -f (p)-

ppXpYpZp ^ '^^GpppXp^^YpZp

This enzyme was a generous gift from
Dr. Bernard Moss at NIH; he and his
co-workers have shown it has an absolute
requirement for 5' diphosphate or tri-
phosphate termini. It will not cap mono-
phosphate or hydroxyl-terminated mole-
cules. On the well-founded assumption

Bottom

Fig. 62. Size of RNA bearing diphosphate or
triphosphate termini from nucleoli of X. laevis
oocytes. Nucleoli were isolated from immature
oocytes of X. laevis, dissolved in SD8, and their
RXA fractionated on a sucrose gradient. Pooled
fractions from the gradient were then tested for
their relative ability to accept 5' terminal caps,
using 'H S-adenosyl methionine as the label
donor.

that diphosphate or triphosphate termini
mark the site of RNA chain initiation,
this enzyme complex allows us to label
selectively the 5' ends of primary
transcripts.

Figure 62 shows an experiment in
which RNA from the nucleoli of Xenopus
oocytes (see Year Book 75, p. 24, for
isolation procedure) was fractionated on

1000

ro

500-

-2

-4 -5 -6 -7

I I I W I

a)

-2 -3 -4 -5 -6 -7

Ml III

RNase T

c)

10 20 30 40 50

Fraction no.

Fig. 63. Digestion of capped high molecular
weight nucleolar RNA with various RNase.
The HMW region of the gradient shown in
Fig. 62 was pooled and capped using ^H S-
adenosyl methionine as the label. The capped
RNA was then digested with various RNase,
and the resulting oligonucleotides were frac-
tionated on columns of DEAE-Sepliadex (0.6 X
25 cm, gradient 0.1 M-0.5 M NaCl in 7 M urea,
0.01 M Tris-HCl, /^H 8). (a) RNase Pi, (b)
RNase T., (c) RNase T,, (d) RNase A. The
elution position of unlabeled marker oligonu-
cleotides of known charge is indicated for each
column profile.

DEPARTMENT OF EMBRYOLOGY 89

a sucrose gradient, and various fractions cuts after every base except when the 2'
were tested for their ability to be capped, hydroxyl on the ribose is methylated.
In this expei*iment about 22% of the 40aS This fact, coupled with the observation
ends were able to be capped and there- that RNase A (cuts after C and U resi-
fore had a 5' diphosphate or triphosphate dues) does not cut within the Tl frag-
terminus. ment, allows us to deduce that the Y base
Figure 63 shows several experiments is A. Therefore, the 5' end of the un-
that were done to characterize the capped capped 40>S molecule begins fplppXp-
40/S termini. In Fig. 63a the capped RNA ApGP. . . . More sequence data on the W
was digested with RNase PI, a nuclease end of the 40*S should be obtained soon,
that cleaves after every base, is not As further proof that we are capping
blocked by 2'0 methylation, and leaves the 5' end of the 40*S molecule, radio-
s' phosphate termini. Chromatography actively capped molecules were hy-
on DEAE yielded a single peak with a bridized to restriction fragments that
charge of about — 2.5. This charge and span the entire rDNA repeating unit,
the specificity of the PI nuclease is con- Figure 64 gives a map of the rDNA
sistent with a structure of ^^OpppX"". repeat showing the location of the re-
Such a result is good evidence that all striction fragments we used. Figure 65a
of the ^H methyl label is actually going shows that 85% of the labeled termini
into cap structures and not into internal which hybridized to rDNA hybridized to
sites along the RNA chain. Digestion the fragment II, the fragment which
of capped 40*S RNA with RNase A and spans the W end of the 40*S sequence,
subsequent DEAE chromatography (Fig. Furthermore, when RNA was recovered
63d) yields one large peak of greater from the hybrid and digested with RNase
than —7 charges and several smaller A, it yielded the two large oligonucleo-
peaks. As we will show below, only the tides that were obtained from unhybrid-
two peaks with greater than —7 charges ized 40*S RNA (compare Figs. 65b and
come from 408 RNA; the two peaks with 63d). This result confirms that the large
less than — 7 charges have not been majority of RNA initiation events on the
studied further. Redigestion of the large rDNA occur at or near the W end of the
RNase A fragment with RNase Tl yields 40*S sequence. It also suggests that the
a single fragment with — 6.5 charges 5' end of the 405 may be heterogeneous.
(Fig. 63c). The identical result is ob- The nature of this heterogeneity requires
tained when intact 405 RNA is Tl- further study.

digested. Since Tl cuts only after G About 15% of the hybridization seen

residues, we can deduce that the struc- in Fig. 63a was associated with fragment

ture of the Tl fragment is ^^GpppXp^- I, a fragment from the middle of the

YpGp. Figure 63b shows that digestion 405 sequence. RNase A digestion of this

of capped 405 with RNase T2 yields a hybrid yielded a single large oligonucleo-

fragment with about — ^.^ charges. T2 tide of greater than — 7 charges (Fig.

5' 40S 3'

^

28S Spacer tsp I8S tsp 28S

^— — ^■^— 4

■^— IV-H II 1 I 1

Fig. 64. Map of one repeating unit of X. laevis rDNA. This map shows the location of frag-
ments I, II, III, and IV which were used for the hybridization experiment shown in Fig. 65.
Fragments were derived from clones Xlrll and Xlrl2 using BamBl and EcoRl.

90

CARNEGIE INSTITUTION

C

E

(/)
o

X

800

600-

400-

200-

20 30 40

Fraction no.

c
E

(/)

c3

X
ro

100 -

100 -

-2

t

~3 -4

-5 -6 -7

i t I

■2 -3 -4-5-6-7
i + t tl +

^^ • ^ *ir*^ >'•>»'

10 20 30 40

Fraction no.

b)

c)

50

60

Fig. 65. Hybridization of capped nucleolar RNA to rDNA restriction fragment. ^H capped
nucleolar RXA (HMW fraction from Fig. 62) was hybridized to restriction fragments im-
mobilized on a nitrocellulo.se filter. The map location of the fragments is shown in Fig. 64.
After the hybrid was trimmed with RNase A, the filter was cut in pieces and counted. Hybrid
RNA was then removed from the filter pieces, digested with RNase A, and the labeled
oligonucleotides fractionated on DEAE-Sephadex as described in Fig. 63. (a) Photograph of
the agarose gel used to .separate the rDNA fragments; underneath is represented the amount
of RNase-resistant radioactivity hybridizing to each fragment; (b) DEAE chromatography of
RNa.se A oligonucleotides recovered from hybrid to fragment H; (c) oligonucleotides recov-
ered from hybrid to fragment I.

65c). The significance of this apparent ping enzymes will be useful for several

initiation event in the middle of the other important purposes. One applica-

gene is unknown. tion will be to sequence the 5' end of the

In addition to locating the 5' end of 40*S RNA. After capping the RNA with

the primary transcript, the Vaccinia cap- a-'^^P GTP, it is possible to partially

DEPARTMENT OF EMBRYOLOGY

91

digest the RNA with base-specific RNase.
Sizing the labeled oligonucleotides on
acrylamide gels should allow rapid se-
quencing of the 5' end. Another applica-
tion will be to study RNA transcription
in vitro. Since we now know the RNase
A oligonucleotide at the 5' end of the 40S
molecule, we should be able to cap RNA
made in vitro and locate correct initi-
ation events in the presence of consider-
able incorrect initiation. Such an assay
will be very useful in attempts to recon-
stitute accurate transcription systems.

DNA Replication Origins in
Xenopus rDNA

Ronald H. Reeder in collaboration with
Robert Benboiv*

Dr. Robert Benbow at Johns Hopkins
recently developed a cell-free system
from Xenopus eggs in which circular
DNA molecules are observed to replicate.
We are collaborating with him to use
this cell-free system to locate the ori-
gin (s) of DNA replication within the
rDNA repeating unit. In preliminary
experiments three types of circular mole-
cules were tested in the system: pXlrll,
which carries the EcoRl fragment from
the middle of the gene region; pXlrl4,
which carries one of the EcoKI spacer-
containing fragments; and CoZEl, the
plasmid vehicle for both EcoRl pieces.
Table 8 shows the percentage of mole-
cules of each type that were converted
into either early or late replicating struc-
tures C'Cairns" forms) after a standard
incubation period. The data suggest that
pXlrll may contain the origin of repli-
cation. Figure 66 is an electron micro-
graph of one of the apparent replicating
structures seen in the pXIrll incubation.
The structures formed from pXIrll
were cleaved with EcoBJ, and the arms
of the resultant branched molecules were
measured. If we assume the molecules
are replicating bidirectionally and that

* Department of Biology, Johns Hopkins
University,

TABLE 8. Formation of Replicating Structures

upon Incubation of Various Plasmids in

Xenopus Ki£ii C!vtoph).'-rn

Plasmid

Percent B Stmcturos*

pXIrll
pXlrl4
ColEl

23

2

* Structures wore scored by electron micros-
copy of 1000 molecules after a standard 4-hour
incubation. Unincubated samples give no repli-
cating structures.

both forks travel at the same rate, the
origin of replication maps to a point in
the rDNA insert about 20% of the insert
length from one of the ^coRI sites. This
would put it either in the internal tran-
scribed spacer or in the 2SS gene region.
But if we assume replication is unidirec-
tional (as is usually the case for CoZEl
replicating in E. coli) , then the origin
maps out in the ColEl vehicle. However,
it does not map to the known location of
the in vivo Coffil origin of replication.
These preliminary results suggest that the
origin may be within the transcribed gene
region.

In sum then, the system looks promis-
ing and offers the hope that further work
will enable us to definitely locate the
origin of replication of these genes.

RiBOSOMAL Gene Chromatin

Harvey L. Wahn

The extrachromosomal nucleoli from
Xenopus oocytes can be isolated free of
bulk DNA contaminants, using pro-
cedures described in Year Book 75 (p.
24). As we began to characterize the
rDNA chromatin in these nucleoli, it
became apparent that proteolysis was a
problem and that further improvements
were needed in the purification scheme.
Previously, two consecutive Metrizamide
buoyant density gradients were required
to purify the nucleoli. Instead, we now
layer the oocyte homogenate over a single
preformed gradient. The interface is

92

CARNEGIE INSTITUTION

Fig. 66. Electron micrograph of a replication structure formed from pXlrll. Circular
molecules of pXIrll were incubated 4 hr in egg cytoplasm fractions under standard conditions.
The DXA was then extracted and spread for microscopy.

stirred to prevent sharp discontinuities in
density. Nucleoli are then sedimented to
their equilibrium position in the gradient,
and soluble protein and less dense orga-
nelles are left behind. This procedure not
only yields nucleoli of high purity but
rapidly removes them from soluble en-
zymes that cause variable recovery of
chromatin proteins. AVe also improved
recoverj^ of protein by adding phenyl-
methanyl-sulfonylfluoride to all prepara-
tions and adding diisopropyl-phospho-
fluoridate when polymerase activity was
not being studied. After recovery from
the gradient, nucleoli were centrifuged

through 0.3 M NaCl to remove loosely
bound protein.

Oocyte nucleoli consist of a core re-
gion containing the rDNA and associated
proteins surrounded by a variable amount
of cortex composed of RNP particles in
various stages of processing. Previous
experiments showed that intact nucleoli
were heterogeneous both in buoyant den-
sity and in RNA chain elongation ac-
tivity in vitro (Year Book 75, p. 24). We
have done experiments to see if this
heterogeneity persists after removal of
the RNP cortex. Isolated nucleoli were
dissociated by suspending them in 10

DEPARTMENT OF EMBRYOLOGY

93

mM Tris, pH 8, 1 mM EDTA and then discontinuities in density and then centri-

layering them over a preformed Metriza- fuged to equilibrium. Fractions from the

mide density gradient. The top of the gradient were then assayed for rDNA

gradient was stirred to remove any sharp and endogenous RNA polymerase activ-

2 3 4 5

7 8

10

12 i3 14 15

130

68

60

- 43

25.7

3000

~ 2000

E

m

^ !000

m

* ■ujfe?. iT:rt0.rr...T»»rW- '

10

?

I

<

Q

>
o

DC

Fraction no.

Fig. 67. Metrizamide buoyant density gradient of nucleolar cores. Intact oocyte nucleoli
(isolated by a previous Metrizamide gradient) were dissociated by resuspension in 10 m3/
Tris pH8, 1 mM EDTA and spun to equilibrium in a second Metrizamide gradient. Aliquots
from each fraction were assayed for endogenous RNA polymerase activity (•-•-•) and
relative rDNA content (0--0--0), and their proteins were electrophoresed on a lO'yc
acrylamide gel in SDS. The bottom (dense) end of the gradient is on the left. Molecular
weights are in daltons ; arrows indicate putative RNA polymerase I subunits.

94

CARNEGIE INSTITUTION

A B C

I 4
I 4

3

2b
2a

mm 4

Fig. 68. Acid-extractable proteins from nu-
cleolar cores. Fractions containing rDNA from
the gradient in Fig. 67 were pooled, extracted
with 0.4 .Y H2SO4, and run on a ISTc acrv'lamide
gel in SDS. CA) acid-extractable proteins from
cytoplasmic ribosomes; (B) acid-extractable
proteins from nucleolar cores; (C) histones
from cultured cells of X. laevis.

ity, and protein from each fraction was
electrophoresed on a 10% SDS-acryla-
mide gel. The results are shown in Fig.
67. About half the protein stayed at the
top of the gradient, while the rest banded
with the rDNA and polymerase activity.

Polymerase activity and rDNA concen-
tration follow each other and no signifi-
cant heterogeneity is observed. Fraction
5 contains three large smeary bands,
indicated by arrows. These proteins do
not follow the concentration of rDNA
and most likely come from a small par-
ticle (melanosome precursor?) that bands
very tightly in fraction 5 and gives that
fraction a grayish appearance. This is
the first experiment in which we have
been able to separate them partially
from the rDNA. Arrows point to bands
in fraction 7 which have the proper mo-
lecular weight to be subunits of RNA
polymerase I. These putative polymerase
subunits appear most concentrated in the
active fractions, except for fraction 5
where the melanosome (?) proteins ob-
scure the relevant region of the gel.

Fractions from the active regions of a
parallel gradient were pooled, and acid-
extractable proteins were electrophoresed
on a 15% acrylamide SDS gel to look
for histones. Figure 68 shows that the
active fraction contains acid-extractable
proteins which electrophorese close to,
but not exactly, with the four nucleosome
histones from bulk chromatin. We are
exploring the possibility that these his-
tones may be modified.

Our working hypothesis is that the
density heterogeneity observed with in-
tact nucleoli results from variations in
the ratio of core to cortex, since removal
of the cortex allows the DNA to band
as a single peak.

Immunological Cross Reaction
BETWEEN Calf and Drosophila Histones

Ronald H . Reeder in collaboration with
Michael Bustin* and Steven L. McKnightf

Histones extracted from calf thymus
are well-characterized proteins whose
primary sequence is known. Antibodies

* National Institutes of Health, Bothesda,
Maryland 20014.

t Df>partmont of Biology, University of Vir-
ginia, Charlottesville, Virginia 22902.

DEPARTMENT OF EMBRYOLOGY

95

against calf histones are available, and
it would be convenient if these antibodies
could be used as universal reagents to
study the arrangement of histones in
chromatin from organisms other than the
calf. To test the feasibility of this, we
compared quantitatively the cross-reac-
tivity between antibodies directed against
calf histones and histones extracted from
Drosophila.

In parallel experiments different dilu-
tions of antibodies against single calf
histones were tested against Drosophila
histones, using quantitative complement
fixation techniques. From the comple-
ment fixation results, an index of dissimi-
larity was calculated for each histone
type. The index of dissimilarity is de-
fined as the ratio of antibody dilu-
tion that gives 50% maximal comple-
ment fixation with homologous protein
to the dilution that gives 50% maximal
complement fixation with heterologous

protein. An index of 1 is obtained when
the homologous and heterologous proteins
are immunologically identical. Higher
indexes indicate a greater and greater
disparity between the two. Indexes of
dissimilarity for calf and Drosophila
histones are shown in Table 9. In several
systems where homologous proteins have
been compared by immunological tech-
niques, a linear correlation exists between
the logarithm of the index of dissimilarity
and the percent amino acid difference of
the proteins compared. Table 9 therefore
also presents the percent amino acid
differences as estimated by this method.
With the exception of HI, which shows
significant species specificity, all of the
Drosophila histones gave very strong
cross-reactions with the homologous calf
thymus histones. The H4 histones in
particular showed no detectable immuno-
logical difference. This agrees with the
fact that Drosophila H4 has the same

TABLE 9. Comparison of Histones Extracted from Drosophila and Calf Thymus

Percent Amino

Percent Difference

Serum

Antigens

Index of

Acid Difference

in Amino

Anti-calf

Compared

Dissimilarity

from I.D.t

Acid Composition^

HI

HI CT*
HID

16.0

27

30 J

H2A

H2ACT
CTMIX

1.35

3

H2A

CTMIX
D MIX

2.0

6

5.0

H2B

H2B CT
CT MIX

2.0

6

...

H2B

CTMIX
D MIX

1.0

9.0

H3

H3
CTMIX

1.7

4

...

H3

CTMIX
D MIX

1.6

4

2.2

H4

H4CT
H4D

1.0

* Abbreviations : CT, calf thymus; D, Drosophila; CT MIX, artificial mixture of calf thj^mus
histones H2A, H2B, H3; D MIX, mixture of Drosophila H2A, H2B, H3. I.D., Index of Dis-
similarity.

t The percent of amino acid difference is calculated by assuming that the published relation
between I.D. and percent of amino acid difference for lysozyme is applicable for histones.

X Taken from published data on amino acid composition.

96

CARNEGIE INSTITUTION

amino acid composition as calf H4 and pected, HI cross-reacts the least. The

the fact that H4 is the most highly con-
served of the histones.

Antisera against calf H2B were also
unable to distinguish between calf and
Drosophila H2B when both histones
were in a mixture with H3 and H2A.

degree of dissimilarity is in good agree-
ment with the known amino acid differ-
ence in HI derived from each of the two
species.

The results unambiguously show that
(with the exception of HI) antisera

Since by amino acid analysis Drosophila elicited by calf thymus histones cross-
H2B shows about 9*^ difference from react very strongly with histones of
calf H2B. this suggests either that the Drosophila. Since the compositional and
amino acid differences are in immuno- electrophoretic differences between calf
silent regions or that they are masked by and Drosophila histones are of the same
association with H2A and H3. Some order as the differences between calf
masking is possible because the H2B histones and a variety of histones ex-
antibody is somewhat more reactive with tracted from other sources, it seems that
calf H2B alone than it is with calf H2B antibodies elicited by calf thymus his-
in a mixture. Slight masking of deter- tones would react strongly with histones
minants due to histone complexing is regardless of origin. We suggest, there-
also observed with H2A and H3. fore, that antibodies for calf thymus his-
Anti-H2A recognizes a small difference tones H4, H3A, H2A, and H2B can be
between calf and Drosophila H2A when used as ''universal" reagents to study the
both are in mixtures. In this case the in situ state of chromatin-bound histones.
immunological difference is almost ex- Histone HI is species specific; and so
actly what would be expected from the for each species, specific antibodies to
known amino acid differences. As ex- this histone have to be elicited.

GENETICS BY GENE ISOLATION:
THE DUAL 5S RNA GENE SYSTEM IN Xenopus

D. D. Brown, J. Doering, N. Fedorojf, L. Kom, E. Jordan, and B. Murr

For several years we have studied a
simple gene system in Xenopus which
is under developmental control. Frogs
synthesize at least two kinds of 5S ribo-
somal RNA. The control mechanism
seems to be that growing oocytes express
all types of 58 RNA genes in the genome,
while somatic cells synthesize a single
kind of 5.S RNA. The goal of our labora-
tory has been to purify and fully char-
acterize the various kinds of 5*S RNA
genes in the genome of more than one
species of Xenopus. We then want to
reconstruct their transcription in vitro
and identify the regulatory molecules
that account for the genes' faithful tran-
scription and developmental control. We
used Xenopus laevis and Xenopus borealis
(misnamed X. mulleri in our earlier

studies) and purified two kinds of 5S
RNA genes from each species.

The major 5S DNA component in each
species synthesizes 5S RNA in oocytes
only (oocyte-type 5S DNA). We have
searched for the somatic-type 5S RNA
genes for some time. A minor 55 DNA
component was detected in the genomic
DNA of each species. The one from X.
laevis was found to code for a 5S RNA
different from both the principal ooctye-
type 5*S RNA and the somatic dS RNA.
We now believe that this is a minor
oocyte-specific 5*S DNA which encodes
10 to 20% of the total 5S RNA accumu-
lated })y oocytes. During the past year
the minor 5S DNA from X. borealis was
identified as the somatic-type 5S DNA.
We now have available for comparison

D.D. Brown and J . B. Gurdon*

DEPARTMENT OF EMBRYOLOGY 97

two 5S DNA's under separate control has shown evidence of transcripts of this

from the same frog. 5*S DNA in oocytes but not in somatic

We present the primary sequence of a cells. We conclude that this is a minor

repeating unit of X. laevis oocyte-type oocyte 5*S DNA controlled in the same

5S DNA. This is the first complete fine- way as the principle oocyte o,S DNA.

structure analysis of a gene and its Additional sequencing, especially of the

spacer that has been determined for an spacer region, will be undertaken soon,
eukaryotic gene. Sequencing is well under
way for the two kinds of X. borealis 5S

DNA. Various homologies have been The Oocyte Injection System

described in control regions of prokary-
otic genes. These are usually detected by
eye. Experience shows how difficult it is Several years ago we began a collabo-

to find even striking regularities in nu- rati ve project to determine whether puri-

cleotide sequences. We expect to search fied genes could be transcribed correctly

for regularities in 5S RNA gene sequences if injected into living Xenopus eggs and

and to compare similar regions in genes oocytes. In the first experiments we in-

from two species or between somatic and jected purified ribosomal DNA (rDNA)

oocyte-type 5S DNA from the same spe- and 5S DNA from Xenopus and mouse

cies. A computer program has been writ- satellite DNA into unfertilized eggs of

ten for this purpose and some of its fea- Xenopus laevis. We detected transcripts

tures are reported here. of the rDNA and 5S DNA but none

A new technique may make it possible whatsoever of mouse satellite DNA. Very

for us to reconstruct proper control of low levels of RNA were transcribed from

5S DNA in vitro. When the purified rDNA and 5*S DNA but were not ana-

DNA is injected into living Xenopus lyzed further at that time. In the spring

oocytes, the oocytes transcribe 5S RNA of 1976 we continued these experiments,

of the correct size from the injected concentrating upon 5S DNA injection

DNA. This permits us to test the faith- into the nuclei (germinal vesicles) of

fulness of transcription of any 5S RNA oocytes. Our renewed optimism was

gene or any fragment or mutant of such based upon experiments of Mertz and

a gene. In addition we hope to begin Gurdon {Proc. Nat. Acad. Sci. U.S.A. 74,

isolation of macromolecules that are re- 1502, 1977) which showed substantial

sponsible for accurate transcription of transcription of SV40 DNA when it was

5S DNA. injected into oocyte nuclei.

All four kinds of purified 5*S DNA

(two each from X. laevis and X. borealis)

Xlt 5*S DNA are transcribed into 5>S RNA of the cor-

^ ^ ^ , ^ , , rect size and sequence by oocyte nuclei

D . D . Brown and E . J ordan ^-r~^^ ^^ ■, _^n m-u t^ivt a j. u

(Figs. 69 and 70). The DNA must be

In Year Book 74 we described some injected into the nucleus of the oocyte;

characteristics of a minor 5*S DNA puri- DNA injected into the cytoplasm is not

fied from the genome of X. laevis (Xlt transcribed. Transcription is judged to
5*S DNA) . This tandemly repeating DNA be mostly faithful because (1) RNA of
contains genes for 5S ribosomal RNA the correct size (Fig. 69) and sequence
which differ in several nucleotides from (Fig. 70) is synthesized; (2) little if any
both the principle oocyte and somatic 5S RNA is transcribed from the anticoding
RNA's. Additional sequencing work has strand of the DNA; and (3) hybridiza-
localized many of these nucleotide sub-
stitutions. Examination of 5S RNA syn- * MRC Unit of Molecular Biologj^ Cam-
thesized by various tissues of X. laevis bridge, England.

98

CARNEGIE INSTITUTION

Ji =

0» =

» =

• =^

-5.8S
— 5S

4S

^ =i

Fig. 69. The synthesis of 5S RNA after injection of purified X. laevis oocyte 5S DNA into
the nuclei of Xenopus oocytes. The DNA was injected for 24 hr followed by [^H]-GTP for an
additional 24 hr. The RNA was extracted and electrophoresed on an acrylamide gel. The
stained gell (right) shows the location of various low molecular weight RNA's present in the
oocyte. Radioactive RNA was detected by fluorography (left). Slots 1 and 3, DNA injection;
slots 2 and 4, buffer injected without DNA.

tion analyses showed that spacer tran-
scripts are much rarer than gene tran-
scripts. The fact that the injected DNA
was the template for transcription was
confirmed bv fingerprinting the [-"^^P] 5S
RXA. In afl cases the labeled 5*S RNA
had the fingerprint expected from the
known sequence of the 5»S DNA used for
injection.

High molecular weight 5>S DNA iso-
lated from the frog as well as cloned
repeating units have been used for these
experiments. The high molecular weight

55 DNA from the frog was found to
support dS DNA transcription for as long
as several days after injection. There is
a lag of 30 to 60 minutes after the DNA
is injected and before 5S RNA is syn-
thesized. If the DNA is injected before
the isotope, then substantial 5S RNA
synthesis can be detected in the shortest
labeling time that we have tried — 10
minutes. This experiment is important
because it confirms previous studies
which showed that there is no precursor
of 5*S RNA.

DEPARTMENT OF EMBRYOLOGY

99

19 •
17'

10 •

0^

8

6

22 8

6

7

4 5

21

Fig. 70. A fingerprint of radioactive 5S RNA synthesized by Xenopus oocytes after the
injection of X. borealis somatic 5S DNA. The X. borealis 5S DNA purified from genomic DNA
of the frog was injected into X. laevis oocyte nuclei along with a-[''T]-GTP. The radioactive
5>S RNA was purified on a gel like the one shown in Fig. 1 and was digested with RNase Ti.
The two-dimensional fingerprint of this digest (right) is compared with [^^P]-5/S RNA purified
from somatic cells of X. borealis (left). The identity of the two fingerprints proves that the
DNA component contains the genes for somatic 5S RNA.

A variety of cloned 5*S DNA repeats
support 5aS RNA synthesis after their
injection into the germinal vesicle. Single
repeating units of 5*S DNA are tran-
scribed into 5»S RNA molecules of the
correct size. This finding demonstrates
that one repeating unit contains the con-
trol sequences required for faithful tran-
scription. In addition we conclude that
methylation of 5S DNA as it is isolated
from the frog is probably not required
for faithful transcription. These cloned
repeating units have been grown in bac-

teria for many generations, and this has
not altered their ability to support 5S
RNA synthesis in the oocyte system.

The frog oocyte system makes it possi-
ble to test a repeating unit for its ability
to support correct transcription of 5S
DNA. We now have an assay system for
detecting mutations affecting control re-
gions in 5*S DNA. The oocyte nucleus
clearly contains excess molecules capable
of regulating transcription of added 5S
DNA. We hope to be able to isolate these
molecules.

100

CARNEGIE INSTITUTION

Sequencing Studies on the 5S DNA of
Xenopus laevis

N. Fe dor off

Using direct DNA sequencing tech-
niques, we have assembled a general pic-
ture of Xenopus laevis oocyte 5*S DNA
at the nucleotide sequence level. We have
sequenced end-labeled restriction enzyme
fragments derived both from total frog
bS DNA and from single repeating units
cloned in bacterial plasmids (see Year
Book 75, pp. 12-21 ; and Maxam and
Gilbert. Proe. Xat. Aead. Sci. 74, 560,
1977). The organization of the repeating
unit is shown in Fig. 71, top. The re-
peating unit, which averages 700 nucleo-
tides in length, can be divided roughly
into an AT-rich half (region A) and a
GC-rich half (region B). The AT-rich
spacer differs in the length of repeating
units, most commonly by multiples of
15 nucleotides. The GC-rich region,
which is the same length in different
repeating units, includes the gene and a
related sequence termed the pseudogene
fJacq, Miller, and Brownlee, 1977; in
press).

Figure 71 shows a composite sequence
for the noncoding strand of the repeating
unit in oS DNA. It represents the domi-
nant sequence and some of its variants
in the oS DNA population. Most of the
sequence is almost identical in a majority
of repeating units. The principal way in
which repeating units differ from each
other is in the redundancy of the oligo-
nucleotide CAAACTTTGAGTTTT at a
single site commencing 41 nucleotides
from the left end of the AT-rich spacer.
Among the six cloned fragments ana-
lyzed, the number of copies of this oligo-
nucleotide varied from 2 to 12. The
modal value for its redundancy in the
total population of repeating units is 5.
We have designated the region contain-
ing this oligonucleotide region A2. It is
indicatf^d by a single copy of the redun-
dant oligonucleotide enclosed in square
brackets; its variable redundancy is
designated by the subscript n. Apart

from region A2, the AT-rich spacer se-
quence is the same in a sufficiently large
fraction of the repeating units in the
population so that a unique sequence
can be determined in total 5*S DNA from
either end of the spacer. Analyses of
cloned single repeating units of 5S DNA
reveal in the relatively constant portions
of the spacer (regions Ai and yla) at
least two types of heterogeneity: base
changes and short deletions. Of the six
fragments examined, no two are exactly
alike throughout the spacer. In addition
to differences in the variable region {A2)
most fragments differ by one or a few
single bases elsewhere in the AT-rich
portion of the spacer {Ai and ^3). Two
cloned fragments lack the oligonucleo-
tides underlined in Fig. 71; one lacks
two short oligonucleotides near the ends
of region A^ (broken line), and the other
lacks a single long oligonucleotide in
region ^3 (solid line over sequence).

The GC-rich portion of the repeating
unit contains the gene and a related se-
quence, the pseudogene. The pseudogene
was first identified, mapped, and se-
quenced by Jacq et at. It commences at
a distance of about 73 nucleotides from
the right of the gene and strongly re-
sembles the first 100 nucleotides of the
gene. These sequences are aligned ver-
tically in Fig. 71. The nucleotides that
differ between gene and pseudogene are
underlined. Beyond nucleotide 100 of the
pseudogene, the sequence resembles the
AT-rich spacer. We have designated nu-
cleotide 106 the 3' terminus of the
pseudogene. Our studies on this sequence
in total frog DNA substantiate those of
Jacq et al., with the exception of pseudo-
gene nucleotide 15. Here we find a T
residue in the noncoding strand, while
they find a C residue. The sequence of
slightly more than 70 nucleotides sepa-
rating the gene and pseudogene (region
B2) resembles a region of equivalent
length immediately on the 5' side of the
gene. These regions are aligned in Fig.
71 to emphasize their similarity. Al-
though the sequence of region B2 is not

DEPARTMENT OF EMBRYOLOGY

101

Hind m

-M-

Haein HaelH

_J_J_

Hindnr
Hoe HI

/ \ AT-rich spocer i ,5S gene i ipseudogene

Region A

Region B

Spacer

5S gene

Pseudogene

Hindu

CAAAGCTTCATTTTTTCAAGGTTTGATTTTTTAAAGTTTT[cAAAGTTTGAGTTTT]^CAAAgTTTICAAAGTTTAATTTTTCAAAGTTTTCAACGTTTTCAAA^TTTGATTTTTCAAAG
TTTTCAAAGTTTAAATTTTTTCAAAGTTTTCAAAGTTTGATTTTTTCAACGTTTTCAAGGTTTGATTTTTCAACGTTTTCAAAGTTTCATTTTTttAACGTTTTCAACGTTTTCAAGGTT

-lOOg -80g

TGATTTTTCAACGTTTTCAAAGTTTCATTTTTCAGTTTTCAGTTTCATTT

Region A

-60Q -40g -209

TTCAAAGTTTTCATTTTCATTTTTCCACAGTGCCGCTGACAAGTCAAGAAGCCGAAAAGTGCCGCTGTTCATC

Region B

Haem

19 I

Haem

60g I

20g 40g 60g 1 80g lOOg I20g

GCCTACGGCCACACCACCCTGAAAGTGCCTGATCTCGTCTGATCTCAGAAGCGATACAGGGTCGGGCCTGGTTAGTACCTGGATGGGAGACCGCCTGGGAATACCAGGTGTCGTAGGCTT

-60P -40p -20p

I I I

TTCAAAGTTTTCAACTTTATTTTGCCACAGCATCGCCGACAAGTCATGGAGCCAAAAG6TGCTGCTGTTCATC

20p

40 p

Haem
60p X

80p

lOOp

GCCTATAGCCACACTACCCTGAAAGTGCCTGCTCTCGTCTGATCTGTGAAGTGATACAGGGGCAGGCCTGGTTAGTACCTGGATGGGAGACCGCCTGAGAAGTTTJCAAAGCTT)

] t

Hindm

Fig. 71. A composite sequence of the repeating unit in Xenopus laevis 5S DNA represent-
ing the dominant sequence variants. The noncoding strand is shown. The total sequence is
subdivided into regions as shown on the diagram. Sequences are numbered with respect to the
first nucleotide of the gene (e.g., lOg is the tenth nucleotide of the gene, while — lOg is the
tenth nucleotide preceding the gene on the 5' side in the sense of transcription), the pseudogene
(similarly, lOp or — lOp), and the left end of the AT-rich spacer (e.g. 10a). The T residue
enclosed in brackets designates a T cluster of uncertain length, while the position at which two
residues appear indicates high levels of heterogeneity at that nucleotide. In the sequence ex-
tending from the left end of region A3 to nucleotide — 120g, the A and T cluster lengths were
assigned from published RNase Ti oligonucleotide data derived from spacer transcripts (Bro"v\Ti-
lee et ah, J. Mol. Biol. 89, 703, 1974) and are therefore somewhat tentative. The underlined
oligonucleotides in region A3 were missing from one of the six cloned 5S DNA fragments
analyzed, while the overlined oligonucleotide was missing from another. Region A2 is repre-
sented by a single copy of the most common variant of the redundant 15-nucleotide sequence
enclosed in square brackets. Its variable redundancy is indicated by the subscript n; n varied
from 2 to 12 in the six cloned repeating units analyzed, and its modal value in the total popula-
tion is about 5. The variants CAAAGTTCGAGTTTT, CAAAGTTCGAGTATT and
CGAAGTTCGAGTTTT have been observed both in cloned fragments and in the total popu-
lation of 55 DNA spacers. Region Bi and the adjacent portion of region A3 are aligned ver-
tically with region B2 two lines below it to emphasize homology. Nucleotides in region B2 which
differ from the corresponding ones at the right end of A3 and in Bi are underlined. The
bracketed regions containing an X designate uncertain portions of B2; the subscripts indicate
that the uncertain portions are 6-7 nucleotides long. The gene is similarly aligned with the
pseudogene two lines below it. Again, nucleotides in the pseudogene that dififer from their
counterparts in the gene are underlined.

complete, the resemblance is already the entire sequence extending from nu-
striking. Thus the GC-rich portion of cleotide —73 on the 5' side of the gene
the repeating unit appears to duplicate through nucleotide 100 of the gene.

102

CARNEGIE INSTITUTION

We have determined the sequence for
the GC-rich portion of the repeating unit
primarily in total frog DXA. As does
most of the AT-rich spacer, frog DNA
gives a unique sequence, which is there-
fore the dominant sequence in the popu-
lation. For the gene, the dominant DNA
sequence corresponds to the most abun-
dant 58 RXA sequence. The sequence
was originally described by Wegnez,
Monier, and Denis [Febs. Lett. 25, 13,
1972). Some portions of the GC-rich re-
gion have been examined in cloned frag-
ments. Where this has been done, the
sequence obtained is identical to the
dominant sequence in total frog DNA.
Oocyte 5S RNA is itself heterogeneous;
two alternative bases have been detected
at seven positions (Wegnez et al., 1972).
Judging from base changes that affect
restriction enzyme sites, we believe that
similar levels of heterogeneity exist in
the DNA.

A striking feature of the primary se-
quence in dS DNA is its hierarchical
redundancy. The AT-rich portion of each
repeating unit consists of the basic short
repeating units CAAAGTTT(T) and

G

CAT5-6. These are arranged in longer

A

internally repetitive patterns, which are
repeated within the spacer. The GC-rich
portion of the spacer, though more com-
plex, is also internally redundant. Finally,
the family of repeating units consists of
a fairly constant arrangement of the
repetitive elements repeated many thou-
sands of times. The variation in length
observed from repeating unit to repeating
unit is related to the internal redundancy
of the repeating unit. That is, the most
variable region of the repetitive unit is
an almost perfectly redundant portion of
the AT-rich spacer (region A2). This
variable region contains, on the average,
five tandem copies of the oligonucleotide
CAAAGTTTGAGTTTT. Next in fre-
quency are short deletions elsewhere in
the AT-rich portion of the spacer (region
^4.3). Except for those in the variable re-
gion, we cannot be absolutely sure that

the observed changes in cloned spacer
fragments correspond to spacers occur-
ring in the frog 5»S DNA population,
especially in view of the instability to
cloning of some satellite DNA's (Brutlag
et ai, Cell, 3, 509, 1977). However, it is
likely that there are such spacers in the
frog 5S DNA because many of the re-
peating units fall outside the regular in-
cremental classes associated with changes
only in the length of the variable region
An. Finally, the GC-rich portion (region
B) shows the lowest redundancy level
within the repeating unit and the great-
est length constancy in the population.
There is a striking discontinuity in the
spacer sequence 50 nucleotides preceding
the gene on its 5' side in the sense of
transcription. To the left of this point,
the spacer is highly repetitive and rich
in A and T clusters. The 49-nucleotide
sequence adjacent to the gene (region
5i) has a much higher GC content and
lacks the obviously repetitive character
of the rest of the spacer (Fig. 72). It
contains several more or less perfectly
repeated oligonucleotides of no longer
than 10 residues. All of them are quite
different from the repetitive units in the
AT-rich portion of the spacer. This
anomalous region also contains several
palindromes, symmetry elements not
found elsewhere in the spacer, but no self-
complementary sequences. Since the 5*S
gene is itself the transcription unit, we
believe it likely that the region adjacent
to the gene on the 5' side contains se-
quences required for correct initiation
of transcription.

The Structure of X. horealis
Oocyte and Somatic 5S DNA's

Jeffrey Doering

In Year Book 75 (p. 14) we reported
the preliminary restriction enzyme map
of one cloned Hindlll fragment of X.
horealis oocyte 5*S DNA (Xbo3). This
map and one for another fragment
(Xbol) have been completed (Fig. 73)

DEPARTMENT OF EMBRYOLOGY

103

10 20 30 kO

4' i + 4-

CCACAGT6CCGCTGACAAGTCAAGAAGCCGAAAAGTGCCGCTGTTCATC

The repeated sequences are:

The symmetrical regions are

5

AGTGCCGCTG

14

3

ACAGT GC

9

34

AGTGCCGCTG

43

17

ACAGTCGC

10

1.000

0.875

2E-05

2E-03

11

GCTGACAAGT

20

22

AAGAAGC

28

27

GCCGAAAAGT

36

34

AA AAGC

29

0.800

0.857

3E-03

6E-03

25

AAG CCG

31

27

GCC GAA

32

33

AAGTGCCG

40

40

GCCGTGAA

33

0.750

0.750

7E-03

7E-03

Fig. 72. The nucleotide sequence preceding the X. laevis oocyte 5S rRNA gene is shown. The
homologous regions, their positions, percent matches, and probabilities are listed as they would
be in a typical computer print-out. No regions of dyad symmetries were found. The numbers
before and after each short sequence refer to its position in the total sequence. The number
below the two sequences is the fraction of nucleotides that match perfectly, and the last num-
ber is the probability that this homology would occur by chance (2E — 05 = 2 X 10"^) .

XbO 3

Hind

nr

780

200

190

Hind

m.

730

41 — imw — ii m i iu

190

140 1 I40II 190 1 100

-I--V

/- ^ ^ - M

XbO 1

^

r

H

1 =

I =
I =

Hpa II

Hae HI
Hha I

= GENE

Fig. 73. Restriction maps of two cloned fragments of X. horealis oocyte 5/S DXA. Xbo3 is
2700 base pairs long and Xbol is 800 base pairs. The ends of both fragments are the Hindlll
sites used for cloning. Numbers refer to base pairs. The exact arrangement of 25 base-pair
fragments cut out by Hhal in the middle of Xbo3 has not been determined. The arrows denot-
ing genes point in the direction of transcription (5' to 3') .

104

CARNEGIE INSTITUTION

using standard restriction mapping
methods. Hybridization with oS RNA re-
vealed the location of the gene-coding
sequences in each cloned fragment. It is
apparent that the genes occur in clusters
separated by different lengths of spacer
DNA. Xbo3 contains two gene clusters,
while Xbol has only one cluster. The
number of genes per cluster varies from
one to at least three. All clusters appear
to be organized in the same way, as
shown by the identical location of re-
striction enzyme sites in the gene cluster
regions of Xbol and Xbo3. The distance
between genes in any cluster is a uni-
form 70 base pairs, the same as the gene-
pseudogene distance in Xlo 5S DNA (N.
FedorofT: this Report). Since this spacer
region should contain sequences involved
in transcription termination for one gene
and transcription initiation for the next
gene, we are in the process of determin-
ing its exact nucleotide sequence. We
will then compare these sequences with
the analogous ones already determined
for X. laevis oocyte 5S DNA (Xlo).
Since the spacer sequences of Xbo and
Xlo have diverged considerably, se-
quences that have been conserved may
identify functional transcription control
regions.

The second gene in the Xbol cluster
is missing the Hpall site in the middle
of the gene, and sequence variants in
Xbo 5S RNA indicate that a small per-
centage of the genes lack this site. These
genes also lack the Haelll site at a
distance of about 17 nucleotides from
its 5' end. This site occurs in the other
gene of Xbol and in at least one gene
of Xbo3. Since the resolution of the gels
used in mapping Xbo3 w^as inadequate
to detect this site in genes farther from
the HinrRll end, it may be present in
other genes of Xbo3 as well. In any case,
nucleotide changes such as this one may
prove useful in determining the precise
region and the precise sequence required
for correct transcription, using the oocyte
injection system described above.

The Xbo spacers consist primarily of a

variable number of repeats of a 25 base-
pair sequence cleaved by the restriction
enzyme Hhal (Fig. 74). This sequence
is also being determined.

Thus, in contrast to Xlo, whose repeat-

5' 10' 15' 20' 30' 45'

Fig. 74, Autoradiograph of partial Hhal
digests of Xbo3 run on 3% acrylamide 0.5%
agarose composite gel. XboS was end-labeled
with polynucleotide kinase and then partially
digested with Hhal for the times (in min)
shown above each slot. After electrophoresis the
gel was dried and autoradiographed. The left
end of the spacer is clearly resolved on this gel,
showing the regularly repeating 25 base-pair
Hhal fragment.

DEPARTMENT OF EMBRYOLOGY 105

length heterogeneity is confined pri- there is only one gene in each repeating

marily to the AT-rich spacer, Xbo is unit. The gene-containing fragments cut

heterogeneous in the length of its spacers out by Haelll (58 and 135 base pairs)

and in the number of genes per cluster, have been sequenced by the method of

Since restriction fragments indicative of Maxam and Gilbert (1977) and found

the gene cluster arrangement are found to have the sequence predicted for the

in digests of the whole population of Xbo somatic 5^S gene. The regions adjacent

repeats, the structure seen in Xbo3 and to the 5' and 3' ends of the gene are

Xbol may be characteristic of the total being sequenced for comparison with the

population of repeating units in Xbo analogous sequences of Xbo 5S DXA.

5S DNA. Hybridization studies indicate little or

In Year Book 75, (p. 15) we reported no homology between the spacers of Xbs

the characterization of a new 5S DNA and of Xbo. Whether the spacer of Xbs

from A^. borealis. We have now identified contains a repeating sequence is not yet

it as the somatic-type 5»S DNA (Xbs) known. In contrast to the AT-rich spacers

by several methods. When injected into of Xlo and Xbo, Xbs has a GC-rich

oocytes, as described above, Xbs DNA spacer.

produces a large amount of full-length

5S RNA. A Ti RNase fingerprint of this Nucleotide Sequence Analysis of the

RNA was found to be identical to that Region Preceding the Xenopus borealis

of somatic-type 5S RNA from tissue 5^ rRNA Gene

culture cells (see Fig. 70). Xbs is cut by ^ , ,,

,, ± ' ±' TT- JTTT • 4. Laurence Jay Korn
the restriction enzyme Hinalli into

homogeneous 900 base-pair fragments. As part of our attempt to understand

The restriction enzyme map of one such transcription in eukaryotes, we are se-

cloned fragment (Fig. 75) indicates that quencing the DNA region adjacent to

Xb S 1

Hind Hind

HI m

390 I 66 I 133 I 260 ^44

50 I 135 I 58 I 56 I 54 I 110 | ? I 95 I 72
< 430 ►H 317 >\< 116 — J

j = Hpa n

I =Hae HI

I = Hha I

< — = GENE

Fig. 75. Restriction map of a cloned fragment of X. borealis somatic 5S DNA (Xbsl) . Num-
bers refer to base pairs. The arrow denoting the gene-coding region points in the direction of
transcription (5' to 3'). The question mark indicates a region where the arrangement of Haelll
sites is uncertain.

106

CARNEGIE INSTITUTION

SPACER

PUTATIVE PROMOTER

5S GENE
65---

Hha I
• 60 ^1

t t
Hae Hae
III III

^^7

tt
Hae

III

53

•250-

Hind III

Hpa II

Fig. 76. The restriction enzyme map of part of a cloned fragment of A", borealis oocyte 5/S
DXA (Xbol). The numbers refer to base pairs.

the 5' terminus of the Xenopus borealis
oocyte-type oS rRNA gene. By analogy
with prokaryotic systems, this region
should contain the information for initi-
ating transcription by RNA polymerase.

Restriction enzyme analysis of a
cloned fragment of A', borealis oocyte 5;S
DNA (J. Doering and S. Emmons, per-
sonal communication) reveals a Hindlll/
Hpall fragment of approximately 250
base pairs which contains 65 base pairs
of the dS RXA gene and approximately
190 base pairs of the region preceding
the gene (Fig. 76). The presence of re-
striction cnzvme sites within the sequence
(Hhal and Haelll] see Fig. 76) will
facilitate sequencing studies. By com-
paring the X. borealis oocyte-type 5*S
sequence with the X. laevis oocyte and
A', borealis somatic-type sequences, we
hope to identify specific DNA regions
with regulatory roles. Toward this end,
a computer program has been written to
facilitate analysis and comparison of
nucleic acid sequences (see the report of
Korn et al., below).

DNA sequencing techniques have been
applied to the X. borealis Hindlll/Hpall
restriction enzyme fragment. Results are
still preliminary. Not enough information
is available to compare the putative pro-
moter region with the X. laevis sequence.
However, we have obtained the DNA
sequence from a part of the restriction
fragment which is analogous to the AT-
rich spacer of X. laevis. The X. borealis
sequence, like that of X. laevis, has large

runs of T residues. Specifically, we find
the sequence GTTTTTG repeats with a
periodicity slightly greater than that of
the 15-mer observed in A", laevis 5S
DNA.

Computer Analysis of
Nucleic Acid Sequences

Laurence Jay Korn, Gary L. Queen,*
and Mark N . Wegman^

We have developed a computer pro-
gram to facilitate the analysis of nucleic
acid sequences. Originally designed to
aid our studies on Xenopus 5S DNA se-
quences, the program is of general use
in analyzing and comparing nucleotide
sequences. For example, the program can
examine the sequence of a polynucleotide
(DNA or RNA) for homologous regions
of various types: repeated sequences,
symmetries (sequences that read the
same forward and backward), and dyad
symmetries (sequences that read the
same forw^ard and backward when A is
interchanged with T and C with G). It
can also search two or more sequences
for subsequences they have in common.

The computer program was written in
the programming language PL/1- It con-
sists of 1580 statements organized into

* Dopartmonl of Mathematics, Cornell Uni-
versity, Ithaca, New York 14850.

t Thomas J. Watson Research Center, Inter-
national Business Machines, Yorktown Heights,
New York 10598.

DEPARTMENT OF EMBRYOLOGY 107

40 subroutines. Established structured When none of these conditions hold, the
programming techniques and algorithms comparison is terminated. If the homol-
were used in organizing and writing a ogy thus obtained meets certain criteria
program designed for ease and flexibility set by the user, the homologous regions
of application. One may specify which and their positions are printed.
of 18 independent procedures is to be Some of the criteria that can be set
applied to the nucleotide sequence. Ten are the minimum number of matches,
parameters, which determine in various the maximum allowed length of a loop-
ways the criteria for homology, may be out, and the minimum ratio of correct
set. matches to total length of the homologous

The first procedure of the program regions. The percent of matches is printed

lists the entered sequence with every immediately beneath each homology,

tenth nucleotide numbered. In the sec- The probability of that homology having

ond procedure, the number and percent occurred by chance alone is also calcu-

of each of the four nucleotides in the lated and listed. By specifying that only

sequence is shown. The third procedure statistically improbable matches be

counts each of the 16 possible dinucleo- printed, we can eliminate inferior homol-

tides and lists their frequency. Similarly, ogies while retaining those most likely

other routines display the distribution to be significant.

of trinucleotides, counted in the three We believe the program will be very

reading frames, both separately and helpful in analyzing nucleic acid se-

together. quences. When the 49-nucleotide sequence

Another procedure examines the nucleic preceding the X. laevis 5S DNA gene

acid for regions rich in purines or py- was examined by the computer, several

rimidines. The criterion for richness is potentially interesting homologies were

that six out of eight consecutive nucleo- observed (see Fig. 72). The biological

tide bases be of one type. The program significance of these homologies and their

prints two copies of the nucleotide se- possible role in the control of transcrip-

quence and overlines the purine-rich re- tion await further analysis. Computer

gions in one and the pyrimidine-rich studies of other regions of the X. laevis

regions in the other. A similar routine 5*S DNA and the X. borealis 5S sequence

locates and prints the AT- and GC-rich are in progress, as well as detailed analy-

regions. ses of prokaryotic regulatory sequences.

Another aspect of the program may be
invoked to find all occurrences of a given

short string of nucleotides in the nucleic Preparation and Transcription of

acid sequence. This feature is especially Bromodeoxyuridine-Substituted

useful in locating restriction enzyme sites. Xenovus laevis 5S DNA

The core of our program consists of

algorithms to find repeated sequences, B.L.Murr

symmetries, and dyad symmetries, which Current evidence suggests that differ-

are not perfect. A representative output ential gene expression is determined in

from these procedures is shown in Fig. part by variations of the base sequences

72. To find homologies, the computer in regions adjacent to the transcription

initiates a comparison at each pair of units. To examine this proposition we

nucleotides. The comparison is continued are attempting to change the structure

as long as the nucleotides match, or when of the Xenopiis laevis 5S DNA by sub-

a mismatch does occur, if it is followed stituting bromodeoxyuridine (BrdU) for

by several matches, or if nucleotides can all thymidine residues. There are two

be '^looped out" to find new matches, principal reasons for examining this

108 CARNEGIE INSTITUTION

bromine-substituted DNA. First, it has result of incorporation of the BrdU into

been reported that when the lac-operator DNA.

is fully substituted by BrdU, its affinity Using the oocyte injection system, we

for the lac repressor is enhanced tenfold, can test whether BrdU-substituted DNA

Second, there are several well-documented can act as the template for accurate

examples showing that exposure of differ- transcription of 5*S DNA. The bromine-

entiating cells to BrdU leads to a sup- substituted gene might have a different

pression of specialized proteins. The affinity for the polymerase and the other

evidence suggests that the effect is the proteins necessary for transcription.

THE COLLECTION OF HUMAN EMBRYOS

Ronan O'Rahilly and Ernest Gardner

The embryological collection is housed awaiting publication in the Handbook

at the L^niversity of California, Davis, oj Clinical Neurology, edited by Vinken

and inquiries about the collection should and Bruyn. Work has been begun on an

be addressed to Professor R. O'Rahilly, atlas of the developing human brain,

Carnegie Laboratories of Embryology, with special emphasis on staged material

University of California, Davis Cali- of the embryonic period proper,

fornia 95616. Some one hundred embryos A study of the development of the in-

and fetuses were acquired during the past terpeduncular nucleus in the midbrain

year. of the rhesus monkey and of the human

has been made by Prof. N. J. Lenn

-n, o (Davis), in collaboration with Prof. P.

Developmental Stages IN -o i •. /tt in i tit ^t tt ir

XT -n Rakid (Harvard) and Mr. N. Halfon

Human Embryos /i-^ • \ rru n i • • ^u ^.u

(Davis). The cellular origms, the path-
Work continues on the revision of way of cellular migration, and the early
stages 10 to 23 (R. O'Rahilly), on the structure of the interpeduncular nucleus
tabulation of the timing and sequence of were analyzed. A striking correlation in
developmental events in staged human these events was found between the two
embryos (R. O'Rahilly) , and on the species when staging was used as a basis
establishment of a computer catalogue of comparison, rather than age or crown-
(Alexander Barry). With the aid of Dr. rump length. This work will provide a
J. E. Drumm (Dublin), a comparison firm basis for electron microscopic in-
was made between (1) the Carnegie C.-R. vestigations of this system in the rhesus
length/age data and (2) the information monkey.

now available relating the C.-R. length During the year, Dr. F. Miiller, Uni-

determined ultrasonically in utero and versity of Basel, visited Davis and com-

postovulatory age determined by basal pleted a study of the development of the

body temperature. Very good agreement anterior falcate and lacrimal arteries,

was found. (See Bibliography). Drs. Miiller and O'Rahilly are investi-
gating the early development of the

T>, TVT ci capuchin monkey, Cebus, and have pre-

Development OF THE Nervous System ^ , ,, -^ I i i - -n . ^ xi_

pared a cardboard model to illustrate the

Work continues on the developmental development of the human pancreas.

neurobiology of primates (E. Gardner Prof. W. Wozniak, Medical Academy of

and associates). A chapter on "The de- Poznan, visited Davis and continued his

velopmental anatomy and histology of studies of the developing brain and spinal

the human central nervous system" is cord. An article by Wozniak and associ-

DEPARTMENT OF EMBRYOLOGY

109

ates on the types of neural cells in the
spinal ganglia of human embryos and
early fetuses has been submitted for pub-
lication. It includes electron microscopic
as well as light microscopic data. An in-
vestigation of the onset and sequence of
joint formation and ossification in the
presacral portion of the human vertebral
column is being undertaken by Professors

A. Ornoy (Jerusalem), E. Gardner, and
D. J. Gray (Stanford).

Visitors to the collection included Pro-
fessors E. C. Amoroso (Cambridge) , P. A.
de Vries (San Jose), G. Gitlin (Stanford
and Jerusalem), H. N. Marvin (Little
Rock, Arkansas), W. P. Luckett (Oma-
ha), B. Maron (Bethesda, Maryland),
and E. M. Ramsey (Washington, D.C.).

STAFF ACTIVITIES

During the year staff members partici- versity ; University of Maryland ; Rocke-
pated as chairmen or speakers at the feller University; National Jewish Hos-
following conferences: Cold Spring Har- pital, Denver; National Institutes of
bor Symposium on '^Chromatin"; Gordon Health; University of Texas; Duke Uni-
Conferences on '^Animal Cells and Vi- versity; University of Maryland, Balti-
ruses," on "Chromatin," and on "Lipid more County Campus ; University of Ala-
Metabolism"; Rothschild School on bama Medical Center; Case Western Re-
"Membrane-Cytoskeleton Interactions" serve; Cold Spring Harbor Laboratories ;
in Jerusalem; Membrane Biology Col- State University of New York at Stony
loquium at Rockefeller University ; Neu- Brook and at Albany; University of
rosciences Program Symposium at Wash- Connecticut Health Center ; Florida State
ington University; Neurosciences Re- University; Roswell Park Memorial In-
search Program Work Session on "The stitute; University of Pittsburgh; Uni-
Developmental History of the Neuron" ; versity of Chicago ; Northwestern Univer-
Tissue Culture Association Decennial sity; University of Indiana; University
Review Conference on "Gene Expression, of North Carolina; Roche Institute of
Regulation and Cultured Cells"; Muscu- Molecular Biology; St. Louis University;
lar Dystrophy Association Fifth Inter- Princeton University ; and Massachusetts
national Scientific Congress; NIGMS Institute of Technology.
Medical Training Program Workshop on Beside the participation of staff in
"Current Problems in Neurobiology"; international symposia, they presented
National Paraplegia Foundation on lectures at the MRC Unit of Molecular
"Growth of the Neuron"; Meeting on Biology, Cambridge, England; ISREC,
Biomolecular Evolution, Paris; Work- Lausanne; Hebrew University, Jeru-

shop on "Silk Gland Differentiation,"
Lyon; Tenth International Congress of
Biochemistry, Hamburg; M. D. Anderson
30th Annual Symposium on Fundamental
Cancer Research: "Cell Differentiation
and Neoplasia"; Plenary Lecture at the
American Society of Biological Chemists;

salem; Universite Libre of Brussels;
Friedrich-Miescher Laboratorium in
Tubingen ; and the Max Planck Institute
at Munich.

Members of the staff served on study
sections of the National Institutes of
Health in Cell Biology and in Bio-

Symposium on "Genetic Regulation in physics and Biophysical Chemistry, and

Eukaryotic Cells," Washington Univer- on the National Science Foundation's

sity; EMBO Workshop on "Sequences in section on Cell Biology.

Genes and Spacers," Davis; and Sym- Editorial service for journals was per-

posium on "Developmental Biology," formed as the Editor of Developmental

Indiana University. Biology and Associate Editor for the

Lectures were given at Harvard Uni- Journal of Cell Biology and Cell.

110

CARNEGIE INSTITUTION

A staff member assisted in the design
of the Hall of Evolution at the Smith-
sonian Museum of Natural History.
Additional consultant activities included
membership on visiting committees for
the Department of Biology at ]\Iassa-
chusettc? Institute of Technology, the De-
partment of Biochemistry and Molecular
Biology at Harvard University, and the
Roche Institute of ^Molecular Biology,
and service on the Board of Scientific
Counselors of the National Institute of
Child Health and Development.

Six staff' members hold joint appoint-
ments in either the Department of Bio-
physics or the Department of Biology at
Johns Hopkins University; they assisted
in courses on advanced cell biology, de-
velopmental biology, advanced biochem-
istry, and neurobiology.

Continued participation by the staff in
the national debate on recombinant DNA
included the National Academy Forum
on Recombinant DNA, a public debate
at the Smithsonian Institute, the presen-
tation of testimony to the Environmental
Protection Agency, and several public
debates and lectures in the Baltimore
area.

Seminars

The Department offers at least two
seminars each week, one of them usually

presented by a scientist from outside the
Department. This year 20 invited speak-
ers from Johns Hopkins University and
the University of INIaryland participated.
In addition the following scientists spoke:
Tom Cech, Massachusetts Institute of
Technology; Vincent Chiappinelli, Uni-
versity of Connecticut at Storrs; George
N. Cohen, LTnstitut Pasteur; Tom
Ducibella, Harvard University; Henry
Epstein, Stanford University; Eric Frank,
Harvard University; Werner Franke,
German Cancer Research Center, Heidel-
berg; J. -P. Garel, Roche Institute of
Molecular Biology; Richard Gordon, Na-
tional Institutes of Health; Barbara
Hamkalo, University of California at
Irvine; Wolfgang Hennig, Max Planck
Institute, Tubingen; Tasuku Honjo,
Tokyo University; Ann Hubbard, Yale
University; Jean Lauder, University of
Connecticut; Ted Maden, Glasgow Uni-
versity; Bert O'Malley, Baylor College
of Medicine; Carl Parker, Washington
University at St. Louis; Joan Prives,
Weizmann Institute; Cary Queen, Cor-
nell University ; Bob Roeder, Washington
University ; Shlomo Rotshenker, Harvard
University; K. Scriabin, Harvard Uni-
versity; Virginia Walbot, Washington
University of St. Louis; Sam Ward,
Harvard University; Harold Weintraub,
Princeton LTniversity.

BIBLIOGRAPHY

Brown, D. D., and J. B. Gurdon, High fidel-
ity transcription of 55 DNA injected into
Xenopus oocytes. Proc. Nat. Acad. Sci.
USA. 74, 2064-2068, 1977.

BrowTi, D. D., see also Doering, J. L., Reader,
R.H., andWenauer,P.K.

Bustin, M., R. H. Reeder, and S. L. Mc-
Knight, Immunological cross reaction be-
tween calf and Drosophila histones. J. Biol.
Chem. 2o2, 3099-3101, 1977.

Dawid, I. B., C. K. Klukas, S. Ohi, J. L.
Ramirez, and W. B. Upholt, Structure and
evolution of animal mitochondrial DNA. In
Genetic Function of Mitochondrial DNA,

C. Saccone and A. M, Kroon, eds., Elsevier/
North Holland Biomedical Press, Amster-
dam, The Netherlands, pp. 3-13, 1976.

Dawid, I. B., and P. K. Wellauer, A reinvesti-
gation of 5'— >3' polarity in 405 ribosomal
RNA precursor of Xenopus laevis. Cell 8,
443-448, 1976.

Dawid, I. B., see also Klukas, C. K., Reeder,
R. H., Tartof, K. D., Upholt, W. B., and
Wellauer, P. K.

Devreotes, Peter N., J. M. Gardner, and

D. M. Fambrough, Kinetics of biosynthesis
of acetylcholine receptor and subsequent
incorporation into plasma membrane of

DEPARTMENT OF EMBRYOLOGY

111

cultured chick skeletal muscle. Cell 10,
365-373, 1977.

Devreotes, P. N., see also Fambrough, D. M.

Doering, J. L., S. Emmons, and D. D. Brown,
Characterization of two types of 5S DNA
in Xenopus borealis. Fed. Proc. 36, 878,
1977.

Drachman, D. B., and D. M. Fambrough, Are
muscle fibers denervated in myotonic
dystrophy? Arch. Neurol. 33, 485-488,
1976.

Drumm, J. E., and R. O'Rahilly, The assess-
ment of prenatal age from crown-rump
length determined ultrasonically. Amer. J.
Anat. 148, 555-560, 1977.

Emmons, S., see Doering, J. L.

Fambrough, D. M., Development of cholin-
ergic innervation of skeletal, cardiac and
smooth muscle. In Biology of Cholinergic
Function, A. M. Goldberg and I. Hanin,
eds.. Raven Press, New York, pp. 101-160,
1976.

Fambrough, D. M., Specificity of nerve-muscle
interactions. In Neuronal Recognition, S.
Barondes, ed.. Plenum Press, New York,
pp. 25-67, 1976.

Fambrough, D. M., and P. N. Devreotes, De-
velopment of chemical excitability in skele-
tal muscle. In Biogenesis and Turnover of
Membrane Macromolecules, J. S. Cook,
ed.. Raven Press, New York, pp. 121-144,
1976.

Fambrough, D. M., see also Devreotes, Peter
N., and Drachman, D. B.

Fedoroff, N., P. K. Wellauer, and R. Wall,
Intermolecular duplexes in heterogeneous
nuclear RNA from HeLa cells. Cell 10,
597-610,1977.

Gardner, E., and R. O'Rahilly, The nerve
supply and conducting system of the hu-
man heart, at the end of the embryonic
period proper. J. Anat. 121, 571-587, 1976.

Gardner, E., see also O'Rahilly, R.

Gardner, J. M., see Devreotes, Peter N.

Giza, P. E., see Suzuki, Y.

Gurdon, J. M., see Brown, D. D.

Higashinakagawa, T., H. Wahn, and R. H.
Reeder, Isolation of ribosomal gene chro-
matm. Dev. Biol. 55, 375-386, 1977.

Higashinakagawa, T., see also Reeder, R. H.

Klukas, C. K., and I. B. Dawid, Character-
ization and mapping of mitochondrial ribo-
somal RNA and mitochondrial DNA in
Drosophila melanogaster. Cell 9, 615-625,
1976.

Klukas, C. K., see also Dawid, I. B.

McMahan, U. J., see Muller, K. J.

McKnight, S. L., see Bustin, M.

Meyer, D. B., and R. O'Rahilly, The onset of
ossification in the human calcaneus, Anat.
Embryol. 150, 10-33, 1976.

Miller, Jr., 0., see Reeder, R. H.

Miiller, F., The development of the anterior
falcate and lacrimal arteries in the human.
Anat. Embryol. 150, 207-227, 1977.

Muller, K. J., and U. J. McMahan, The
shapes of sensory and motor neurons and
the distribution of their synapses in ganglia
of the leech: a study using intracellular
injection of horseradish peroxidase. Proc.
R. Soc. Lond. B. 19^, 481-499, 1976.

Ohi, S., see Dawid, I. B.

O'Rahilly, R., Prenatal human development.
In Biology of the Uterus, R. M. W\Tin, ed.,
Plenum Press, New York, pp. 35-57, 2nd
edition, 1977.

O'Rahilly, R., The development of the vagina
in the human. In Morphogenesis and Mal-
formation of the Genital System, R. J.
Blandau and D. Bergsma, eds., Birth De-
fects, Original Article Series, Liss, New
York, 13, 123-136, 1977.

O'Rahilly, R., and E. Gardner, The embryol-
ogy of bone and bones, in Bones and Joints
(Intl. Acad, of Pathology Mono. No. 17),
L. V. Ackerman, H. J. Spjut, and M. R.
Abell, eds., Williams and Wilkins, Balti-
more, pp. 1-15, 1976.

O'Rahilly, R., see also Drumm, J. E., Gard-
ner, E., and Meyer, D. B.

Ozato, K., see Pagano, R. E.

Pagano, R. E., K. Ozato, and J. M. Ruys-
schaert. Intracellular distribution of lipo-
philic fluorescent probes in mammalian
cells. Biochim. Biophys. Acta 4^5, 661-
666, 1977.

Ramirez, J. L., see Dawid, I. B.

Reeder, R. H., D. D. Brown, P. K. Wellauer.
and I. B. Dawid, Patterns of ribosomal
DNA spacer lengths are inherited. J. Mol.
Biol. 105, 507-516, 1976.

Reeder, R. H., T. Higashinakagawa, and 0.
Miller, Jr., The 5'-^3' polarity of the
Xenopus ribosomal RNA precursor mole-
cule. Ce// <?, 449-454, 1976.

Reeder, R. H., see also Bustin, M., Higashina-
kagawa, T., and Wellauer, P. K.

Ruysschaert, J. M., see Pagano, R. E.

Suzuki, Y., Accentuated expression of silk
fibroin genes in vivo and in \\Xyo. Tenth
International Congress of Biochemistry,
Abstracts, p. 77, 1976.

112

CARNEGIE INSTITUTION

Suzuki, Y.. and P. E. Giza. Accentuated ex-
pression of silk fibroin genes in vivo and in
Wtro. J. Mol. Biol. 107\ 1S3-206. 1976.

Tanof. K. D., and I. B. Dawid, Similarities
and differences in the structure of X and Y
chromosome rRXA genes of Drosophila.
Xaturc eeS, 27-^0, 1976.

Upholt, W. B.. and I. B. Dawid, Functional
organization and evolution of animal mito-
chondrial DNA. In Genetics and Biogene-
sis of Chloroplasts and Mitochondria, Th.
Bucher et al., eds., Elsevier/ North Holland
Biomedical Press, Amsterdam, The Nether-
lands, pp. 5S7-592, 1976.

Upholt. W. B.. see also Dawid, I. B.

Wahn, H., see Higashinakagawa, T.

Wall.R...<;rf Fedoroff, N.

Wellauer, P. K., and I. B. Dawid, The struc-
tural organization of ribosomal DNA in
Drosophila melanogaster. Cell 10, 193-212,
1977.

Wellauer, P. K., I. B. Dawid, D. D. Brown,
and R. H. Reeder, The molecular basis for
length heterogeneity in ribosomal DNA
from Xenopus laevis. J. Mol. Biol. 105,
461-486, 1976.

Wellauer, P. K., R. H. Reeder, I. B. Dawid,
and D. D. Brown, The arrangement of
length heterogeneity in repeating units of
amplified and chromosomal ribosomal
DNA. from Xenopus laevis. J. Mol. Biol.
105, 487-505, 1976.

Wellauer, P. K., see also Dawid, I. B., Fedor-
off, N., and Reeder, R. H.

PERSONNEL

Year Ended June 30, 1977
(including those whose services ended during the year)

Research Staff

Donald D. Brown, Director
Igor B. Dawid, Biochemistry
Dougla.^ M. Fambrough, Biochemistry
Kenneth J. Muller, Neurobiology
Richard E. Pagano, Biophj'sics
Ronald H. Reeder, Biochemistry
Yoshiaki Suzuki, Biochemistry

Research Associates (Extramural)

Bent G. Boving, Detroit, Michigan
Robert L. DeHaan, Atlanta, Georgia
Ernest Gardner, Davis, California
Arthur T. Hertig, Boston, Massachusetts
In\in R. Konigsberg, Charlottesville, Vir-
ginia
Ronan O'Rahilly, Davis, California
Elizabeth M. Ramsey, Washington, D.C.

Postdoctoral Fellov:s and
Grant-Supported Associates

Peter Botrhan, Fellow of the U.S. Public
Health Service and Carnegie Institution
of Wa.'>:hington

Salvatore T. Carbonetto, Fellow of the
U.S. Public Health Sorvice

Diana Card, Fellow of Muscular Dystro-
phy Association of America, Inc.

Jeffrey Doering, Fellow of the U.S. Pubhc
Health Service

Mark Dworkin, Fellow of Carnegie Insti-
tution of Washington

Scott Emmons, Fellow of the National
Cystic Fibrosis Research Foundation^

Nina Fedoroff, Fellow of the U.S. Public
Health Service and Carnegie Institution
of Washington; U.S. Pubhc Health Serv-
ice Grant (Brown)

Paul Geshelin, Fellow of the U.S. Public
Health Service^

Elizabeth Godwin, U.S. Public Health
Service Grant (Dawid)

Lawrence J. Korn, Fellow of Helen Hay
Whitney Foundation

Eric Long, Fellow of the Swiss National
Fund for Scientific Research

Yasumi Ohshima, Fellow of Carnegie Insti-
tution of Washington

Keiko Ozato, Fellow of Marine Biological
Laboratory^

Richard Rotundo, Fellow of Carnegie In-
stitution of Washington

Alex Sandra, U.S. Public Health Service
Grant (Pagano)

Barbara Sollner-Webb, U.S. Public Health
Service Grant (Reeder)

Masatoshi Takcichi, Fellow of Carnegie
Institution of Washington*

DEPARTMENT OF EMBRYOLOGY

113

Katherine Tepperman, Fellow of the U.S.

Public Health Service'^
Yoshihide Tsujimoto, U.S. Public Health

Service Grant (Suzuki)
Walter Wahli, Fellow of the Swiss National

Science Foundation
Harvey Wahn, U.S. Public Health Service

Grant (Reeder)
Peter K. Wellauer, Fellow of the National

Cystic Fibrosis Research Foundation^'

Students

Jeffrey Chernak, Undergraduate, Cornell
University

Brian Cooley, Undergraduate, Johns Hop-
kins University

Peter Devreotes, Graduate, Johns Hopkins
University

John M. Gardner, Graduate, Johns Hop-
kins University

Margie M. Goldberg, Undergraduate,
Goucher College

Robert A. Hipskind, Graduate, Johns Hop-
kins University

Les Katzel, Graduate, Johns Hopkins Uni-
versity

Marc C. Krauss, Graduate, Johns Hopkins
University

Jose Ramirez, Graduate, Johns Hopkins
University

Shigeru Sakonju, Graduate, Johns Hopkins
University

Nancy A. Union, Undergraduate, Goucher
College

Visiting Investigators and
Extramural Collaborators

Robert Benbow, Baltimore, Maryland
Michael Bustin, Bethesda, Maryland
Tom Cech, Cambridge, Massachusetts
Michael Edidin, Baltimore, Maryland
Patricia Gearhart, Baltimore, Maryland
John Gurdon, Cambridge, England
Uel J. McMahan, Boston, Massachusetts
Barbara R. Migeon, Baltimore, Maryland
Oscar Miller, Jr., Charlottesville, Virginia
Brown L. Murr, Baltimore, Maryland

Keiko Ozato, Baltimore, Maryland
Cary L. Quf;en, Ithaca, New York
R. M. Ptuysschaert, Brussels, Belgium
Ken Tartof, Philadelphia, Pennsylvania
Hans Weiss, Heidelberg, Germany
Peter Wellauer, Lausanne, Switzerland
John Wienstein, Bethesda, Maryland

Clerical and Technical Staff

Elaine S. Asch, Senior Technician
James H. Blackwell, Custodian
Paul Blackwell, Custodian (part-time)
Jeffrey L. Ciemny, Recorder^
William H. Duncan, Senior Technician
Ernestine V. Flemming, Laboratory Helper
Anne K. Francis, Secretary^
Paul E. Giza, Technician
Richard D. Grill, Photographer
Virginia Hicks, Laboratory Helper
Mary E. Hogan, Technician
WilHam L. Johnson, Custodian
John E. Jones, Custodian
Eddie Jordan, Senior Technician
Catherine R, Lane, Librarian (part-time)
AHce H. Mabin, Laboratory Helper
Thomas F. Malooly, Business Manager
Barbara Melnick, Secretary
Thomas F. Miller, Custodian
Joyce Patterson, Laboratory Helper
John Pazdernik, Building Engineer
Betty Lou Phebus, Bookkeeper
Martha L. Rebbert, Senior Technician
Susan D. Satchell, Secretary
Patricia Schmidt, Secretary
Ginny Selby, Secretary (part-time)
Bessie Smith, Laboratory Helper
Delores Somerville, Senior Technician
Barbara Thomas, Technician
John Wiser, Machinist

' To September 30, 1976.
' To February 14, 1977.
'To August 31, 1976.
* To August 31, 1976.
' To July 29, 1976.
« To Julv 15, 1976.
' To October 29, 1976.
'To April 21, 1977.

Hale Observatories

Operated by Carnegie Institution of Washington
and California Institute of Technology

Pasadena, California

Horace W. Babcock
Director

J. Beverley Oke
Associate Director

OBSERVATORY COMMITTEE

Horace W. Babcock, Chairman

J. Beverley Oke, Vice-Chairman

James E. Gunn

George W. Preston

Allan Sandage

Wallace L. W. Sargent

Maarten Schmidt

Arthur H. Vaughan

Carnegie Institution of Washington Year Book 76, 1976-1977

Contents

Introduction 119

Observing Conditions 124

Physics of the Sun 125

Mount Wilson observations 125

Solar magnetic fields 126

Large-scale velocity fields 127

Solar pulsations 127

Big Bear Solar Observatory 127

Macrospicules, x-ray bright-point

flares and spicules 128

X-ray spectra of flares 128

Physics of flares 129

Solar radio astronomy 129

Solar System 130

lo 130

Comets and asteroids 132

Stellar Spectroscopy 132

White dwarfs 132

Lower end of the main sequence , . 133

B-type stars 133

Red giants in globular clusters 134

Stellar Chromospheres 134

Instrument for the study of

stellar chromospheres 134

Flare stars 135

Millimeter-Band Photometry 135

One-millimeter photometry of

extragalactic objects 135

Thermal sources 135

Nonthermal sources 136

Observing procedures 137

Submillimeter heterodyne

spectroscopy 138

Globular Clusters 138

Infrared studies of red giants in

globular clusters 138

Interstellar Matter and Gaseous Nebulae 139

Planetary nebulae 140

One-millimeter photometry of the

Crab Nebula ' 140

Pulsars 140

Secular decrease in the visible

intensity of the Crab pulsar 140

X-Ray Sources 141

Supemovae 144

The Galaxy: Its Halo and

Globular Clusters 144

Metal abundances in globular clusters 144

M15 main-sequence photometry 145
Color-magnitude diagram of two

remote halo clusters 146

NGC 5053 146

The remote Serpens cluster Palomar 5 146
The outer globular clusters and

satellites 147

Molecular hydrogen in galactic sources 148

Galactic center observations 149

Luminosity function of white dwarfs

and hot subdwarfs 149

H and K objective-prism survey .... 149

Galaxies 150

Composition gradients across spiral

galaxies 150

Binary galaxies 150

Compact galaxies 150

Globular clusters in M31 151

Globular clusters in M87 151

Dynamical studies 152

Velocity dispersions in galaxies 152

Bright spiral galaxies 153

Seyfert galaxies 153

Grouping galaxies 154

Extragalactic observations at

1 to 10 microns 154

NGC 4151 155

NGC 1199 155

Multiplicity of nuclei of ellipticals . . . 156

Clusters of Galaxies 156

"Missing mass" in clusters 156

Velocity field near rich clusters of

galaxies 157

Velocity anomalies in the Virgo cluster 157

Search for evolutionary changes 157

Search for intergalactic hydrogen .... 157

Cluster models 158

Radio Sources 159

3CR identification program 159

Identifications of radio sources 159

Quasars and Quasi-Stellar Objects 159
Photometry and spectroscopy in the

S*^ and 15^' survey fields 160

Bright survey field 161

Spectroscopic observations 161

Infrared photonietry of QSS 162

Absorption-line spectra 162

Peculiar quasars 163

Observational Cosmology 163

The velocity field of nearby galaxies . . 163

Solar motion relative to the

Local Group 164

E\'idence for a positive curvature of
the Hubble diagram for brightest

cluster galaxies 164

The Hubble diagram 166

Theoretical Studies 167

The abundance distribution in the

galactic halo 167

Instrumentation 167

"Charge-coupled" detectors 167

Photon-coimting spectrometers 167

Infrared photometer 168

Palomar aluminizing program 168

5-meter telescope control room 169

Astroelectronics Laboratory 169

Guest Investigators 170

Bibliography 185

Staff and Organization 191

INTRODUCTION

The advance of astrophysics in general
and of cosmology in particular has been
limited by the rate at which data can be
collected. For each investigator who plans
and conducts extensive programs for the
acquisition of basic data, there are sev-
eral who are ready to interpret the ob-
servations and, using them as a founda-
tion, to erect a framework of theory
thereon. In too many instances, over-
extended interpretations or conclusions
have been based on data samples that
are marginal or inadequate. Some im-
provement of this data-limited situation
is resulting from the several large new
optical and radio telescopes that are be-
coming operational in this decade, and
also from new instruments above the
earth's atmosphere. Certainly great im-
provements have already taken place
through the advent of two-dimensional
photon-counting devices for spectrometers
and for image photometry. At the same
time, photographic plates with improved
characteristics remain indispensable for
wide-field recording. When finally it be-
comes possible for astronomers routinely
to measure with high efficiency and high
resolving power the direction, energy, flux
rate, and polarization of incoming pho-
tons over an extended image with negli-
gible contamination from the sky, a
fundamental if not ultimate plateau of
instrumental performance will have been
achieved.

Important progress toward this goal
of improved instrumentation continues
through the work of many individuals.
At the Hale Observatories those most
involved include Gunn, Kristian, Oke,
Persson, Shectman, and Westphal. De-
vices of several types are in use or under
test, but among the most promising are
the CCDs, or charge-coupled devices.
vSome of these are currently being tested
in collaboration with personnel of the
Jet Propulsion Laboratory.

The Observatories have a record of

support of long-term "sustaining" pro-
grams of data acquisition that can pro-
vide a substantial base for many investi-
gators. An example of such a long-term
program is that of acquiring and publish-
ing a consistent body of data on the
observed velocities (redshifts) of all the
brighter galaxies. Now being brought to
completion by Sandage and Tammann,
this effort has been under way for forty
years. It was begun by Hubble in his
contacts with N.U. Mayall of the Lick
Observatory and in direct collaboration
with Milton Humason at Mount Wilson,
beginning in 1936. Its aim was the acqui-
sition of data either by direct spectro-
scopic observation or by compilation
from published sources, to result in red-
shifts of all of the 1249 bright galaxies
of the Shapley-Ames catalogue. (The
catalogue, published by the Harvard Col-
lege Observatory in 1932, lists positions,
angular sizes, and estimated magnitudes.)
A 1956 paper by Humason, Mayall, and
Sandage summarized results of the pro-
gram to that time. New spectroscopic
observations, including many obtained
by Sandage in Australia in 1969 and
others made by him at Palomar from
1970 to 1977, have resulted in 670 new
redshifts for bright galaxies. Now^, Sand-
age and Tammann, in collaboration, have
compiled all known redshifts for the
galaxies in the catalogue. It is expected
that the revision will be published by the
Carnegie Institution in 1978.

The revised Shapley-Ames catalogue
will include reclassification of the types
(spiral, elliptical, irregular, etc.) of all
listed galaxies on the Hubble system.
This reclassification was initiated by
Hubble in 1946 on the basis of improved
direct photographs. Over the years,
Sandage has accumulated plates with the
5-meter Hale Telescope in the north and
with the 1 -meter Swope Telescope at Las
Campanas in the south. These photo-
graphs, together with others in the col-

119

120

CARNEGIE INSTITUTION

lection of plates made with the Blount
Wilson telescopes, form the basis for the
more than 1200 new galaxy types now
being prepared for publication.

The immediate purpose of Sandage and
Tammann in compiling the redshift ob-
servations is to provide a data sample
that will permit determination of the
sun's motion (and of the motion of our
Galaxy) relative to the inner metagal-
axy — "our part of the Universe." Analy-
sis of the new body of data, with its
nearly comj'jlete velocity coverage, has
been begun by A. Yahil of Tel Aviv
University and Tammann with the aim
of measuring perturbations of the local
velocity field in the presence of density
contrasts. The goal is to assess the role
of gravity in determining the global
world model by using the velocity per-
turbations as measures of the local ratio
of kinetic energy to gravitational poten-
tial energy.

In 1962. Eggen, Lynden-Bell, and
Sandage presented observational evidence
showing that our Galaxy formed by con-
traction from a primordial gas cloud.
Stars and globular clusters that we now
identify as ''old" formed during the con-
traction and continue to show a roughly
spherical distribution (the ''halo") and
to pursue inclined orbits around the cen-
troid of the system; these objects do not
interact dynamically with the residual
gas. The remnants of the contracting gas
clouds, however, are impermeable to one
another and do interact, generally col-
lapsing to a plane determined by the
angular momentum of the system. There
they give rise to successive generations
of young stars.

The subject of the structure and evolu-
tion of galaxies is now very active, and
significant new insights are being
achieved. A few years ago, Searle re-
ported the discovery of chemical com-
position gradients across the disks of
spiral galaxies. The discovery has been
sustained, and attention has turned to
the origin of the gradients and to the
detailed physics of the ionized regions

whose spectra provide evidence for these
gradients. In the past year, Searle has
collaborated with G. A. Shields of the
l^niversity of Texas in analyzing new
observations of emission regions in the
spiral IVIIOI. Searle and Shields find that,
in addition to the well-established abun-
dance gradient, there is also a gradient in
temperature of the ionizing stars, the
hottest stars occurring far from the center
of the galaxy. They suppose that the
ionizing stars share the abundance gradi-
ent of the gas and show that this has
important effects on the character of the
ionizing radiation emitted by these stars.
When these effects are both taken into
account, the detailed models produce an
excellent fit to the new observations, and
Searle and Shields find that the 0/H
abundance ratio falls off as r~^/^ when r,
the distance from the center of the gal-
axy, lies between 5 and 25 kiloparsecs.
The observations rule out a theoretical
prediction that the abundance would de-
crease linearly with the radius.

Shectman and Preston have initiated
a search for stars of very low metal
abundance in the halo of our Galaxy. The
discovery of such objects would place
important constraints on the early his-
tory of nucleosynthesis during the con-
traction phase. For this survey, they have
equipped the 46-cm Schmidt telescope at
Palomar with a large interference filter
and an objective prism. The filter limits
the extent of each spectrum to 100 A
around the H and K lines of ionized
calcium. This technique, with a survey
limit near the 14th magnitude, permits
easy rejection of the vast majority of
stars showing normal H and K line
strengths. The residual group of rarer
objects includes white dwarfs, nuclei of
planetary nebulae, and very-low-metal-
abundance stars that are the objects of
the survey.

The use of the SIT Vidicon Cassegrain
spectrograph and the multichannel spec-
trometer has resulted in the discovery of
over fifty additional degenerate stars by
Greenstein and collaborators. They find

HALE OBSERVATORIES 121

continued evidence for stratification in trograph of the Hale Telescope. The
the atmospheres of degenerate stars, and mass-to-light ratio is about 3 in visual
for the presence of relatively sharp but units — rather higher than generally ex-
weak hydrogen lines in the spectra of pected. This has im[)ortant implications
quite cool objects that are dominated by for dark remnants that may contribute
helium. Two red degenerate stars studied to the mass of the cluster,
by Greenstein and Dr. W. F. van Altena Kristian, Sandage, and Westphal are
of Yale University are among the in- continuing their program for measure-
trinsically faintest known, with visual ment of redshifts and magnitudes of
luminosity of about 10~* that of the sun. brightest cluster galaxies to extend the

Miinch has collaborated with Drs. Fred Hubble diagram to large redshifts. They

Roessler and John Trauger of the Uni- now present new photometry for 33

versity of Wisconsin in studying the clusters and new redshifts for 50 clusters,

physics of the magnetosphere of the extending to ^ = 0.75. The new data,

planet Jupiter. They used the PEPSIOS combined with earlier samples, show evi-

spectrometer (a pressure-scanning Fa- dence for a positive curvature of the

bry-Perot interferometer of 150-mm Hubble diagram, with a formal value of

aperture). The instrument offers a re- the deceleration parameter q^ (uncor-

solving power of 130,000 and a circular rected for evolution) of 1.6.

field of view of 30'' in diameter. While To avoid possible selection effects for

many questions remain unanswered, the largest redshift clusters, Kristian,

Miinch has concluded that the heavy-ion Sandage, and Westphal restricted the

abundances approximate those expected analysis of their sample to clusters with

for a meteoritic composition and do not z < 0.4. The currently available data

remotely approach those expected for a samples show, for the first time, a sig-

Jupiter-like composition. The plasma- nificant departure of the Hubble diagram

sphere seems to be chemically isolated from a straight line, with persistent indi-

from the planet and from the solar wind, cation of a positive curvature. If no

The new application of heterodyne evolutionary corrections are applied, the
spectroscopy at submillimeter wave- formal best-fit value for go is 1.6 =b 0.35.
lengths has been begun at the 5-meter The dispersion in the absolute magni-
telescope in collaboration among Drs. tudes of the brightest cluster galaxies
T. Phillips and T. Huggins of the Bell continues to be strikingly small : the best-
Telephone Laboratories and Neugebauer fit values are My = —23.28 and M^ =^
and Werner of the Hale Observatories. — 24.09, with a dispersion of 0.28 mag for
The receiver, based on an InSb hot- the first-ranked cluster galaxy in each
electron bolometer mixer, is mounted at passband.

the prime focus of the telescope. Its use After a lapse of nearly fifty years since
has resulted in the first detection in an Hubble's original discovery of the red-
astronomical source of the 7 = 3 -> J = shift-distance relation, and its extension
2 line of CO at 345 GHz. The line has by a factor of 200 in distance, the data
been mapped over the central few arc appear to be near to fulfilling their
minutes of the Orion molecular cloud, original promise as a possible test for
The emission peak is within 10'' of known cosmological models. The validity of this
infrared sources. approach for finding Qo is now dependent

Gunn and Dr. R. F. Griffin of Cam- on the outcome of efforts to understand

bridge Observatories, England, have com- the nature and size of galaxy-evolution

pleted their observations and analyses of effects during the look-back time. To

individual giant stars in the globular consider two opposite cases, the present

cluster M3 as obtained with their tern- data show that if the universe is nearly

plate spectrometer at the coude spec- empty {qo ^ 0) , then galaxies must have

122

CARNEGIE INSTITUTION

been about 1 mag brighter at 2 = 0.4.
On the other hand, if there has been no
net brightness evohition since z = 0.4,
then the data show that the universe is
iirmly closed, finite, and osciUating.

Wilkinson and Oke have completed a
study of the absolute spectral-energy
distributions of 54 bright<^st galaxies in
faint clusters. About 15 objects have red-
shifts between z = 0.10 and 0.21 and
constitute the sample with short look-
back time. The remaining 39 clusters
have redshifts between 0.21 and 0.47.
Over the range in time corresponding to
z = 0.10 to 0.46, B — V is foimd not
to change by more than 0.03 mag. Other
effects of size comparable to or greater
than any color evolution are present,
and a comparative analysis of the indi-
vidual spectral-energy distributions sug-
gests that one significant effect may be
a dispersion in metallicity by a factor 4
among the galaxies in the sample.

Oke and Richstone used the multi-
channel spectrometer and the SIT-Vidi-
con digit-al spectrograph to resolve spati-
ally the quasar 3C249.1, previously
thought to be stellar. These observations
revealed narrow emission lines of [0 II] ,
[0 III], and [Ne III] from a region
about 2" away from the quasar at a
redshift consistent with that of the
quasar. This is the third instance of such
a phenomenon. It appears to rule out a
gravitational model for the quasar red-
shift.

In general, it has not yet proved pos-
sible to decide where the absorption lines
of quasars and quasi-stellar objects origi-
nate — whether in material ejected at high
speeds from the QSO's or in unrelated,
intervening objects such as galactic halos.
Dr. A. Boksenberg, University College
London, and Sargent used the IPCS at
the coude spectrograph of the Hale Tele-
scope to demonstrate that absorption
lines are produced in the spectrum of at
least one QSO, 3C232 with z,.m = 0.53,
by matfrial in the outer parts of a galaxy,
XGC: 3067. The QSO 3C232 and the Sa
galaxy NGC 3067 are separated by r9

on the sky. The redshift of the galaxy is
z == 0.005. Haschick and Burke (Astro-
phys. J. [Lett.] , 200, L137, 1975) showed
that the 21 -cm line in the radio spectrum
of 3C232 is observed in absorption at a
velocity of 1418 ± 2 km s~^ despite the
fact that the line of sight to 3C232 passes
17 kpc away from the center of NGC
3067. Boksenberg and Sargent observed
the optical spectrum of 3C232 in the
vicinity of the Ca II H and K lines and
at a resolution of 1 A. They found ab-
sorption lines corresponding to Ca II H
and K at a redshift of 1406 dz 11 km s-^.
The lines are quite strong, with an equiv-
alent width of 0.4 A for K. A line of this
strength would require a typical path
length of about 1.3 kpc in the plane of
our Galaxy near the sun; however, it has
been shown that insterstellar lines of
comparable strength can be found in
the spectra of globular-cluster horizon-
^tal-branch stars in the halo of our Galaxy.

The observations by Boksenberg and
Sargent establish that heavy elements
which could produce QSO absorption
lines are present in the interstellar gas
far out in the halo of the disk of a spiral
galaxy. This galaxy presents a much
larger cross section for the production
of absorption lines than would be ex-
pected from its optical size.

A study of the Crab Nebula at a wave-
length of 1 mm has been completed by
Werner, Neugebauer, and associates. The
nebula was mapped with 1' resolution in
twilight observations at the prime focus
of the 5-meter telescope. The peak 1-mm
surface brightness is seen in the center
of the nebula at or near the position of
the pulsar; the distribution of radiation
around this j)eak is roughly elliptical,
with half-power width --2:5 X 4:5. The
result is in agreement with the size and
shape of the nebula as determined at
longer wavelengths. This agreement has
been used to estimate the flux expected
from the faint outer regions. On this
})asis, the total 1-mm flux from the Crab
Nelmla is found to be 300 ±: 80 Jy.

The investigators found that both the

HALE OBSERVATORIES 123

total 1-mm flux from the Crab Nebula special attention to adjustment of electri-
and the spatial distribution of the 1-mm cal, mechanical, and optical systems, and
radiation are consistent with expectations with much effort devoted to completion
from radio wavelength observations. The of essential auxiliary instruments. Ex-
results indicate that the emission from cessive deflection of the support system
the Crab Nebula down to wavelengths as of the Cassegrain secondary mirror had
short as 1 mm can be attributed to syn- to be corrected by designing and install-
chrotron emission from a single distribu- ing a new gravity-sensitive self-aligning
tion of electrons. mechanism in the flip cage. This mecha-

A program of observations from 0.3 nism was adjusted and successfully

to 10 microns of all quasars brighter than tested in February 1977, at which time

F = 17 has been completed with the in- the primary and secondary mirrors of

frared photometer by Becklin, Matthews, the telescope were collimated. In April,

and Neugebauer and by Oke with the the Gascoigne-Bowen correcting lens, the

multichannel spectrometer at the 5- third and final element of the modified

meter telescope. The preliminary results Ritchey-Chretien wide-angle optical sys-

are already clear: (1) There is evidence tem, was finished and installed in the

for a number of types of infrared-con- telescope.

tinuum energy distribution. (2) Although April 1977 saw the first scientific use

the energy distribution cannot be char- of the new telescope, by Dr. Charles H.

acterized by simple power laws, there is, Townes and collaborators for infrared

in general, an increase of flux density investigations of the center of the galaxy

with decreasing frequency. (3) Although and by Babcock and Brucato for direct

some quasars emit most of their energy photography.

in the infrared at wavelengths equal to or Dr. Townes, together with Fred Baas

greater than 10 /a, a significant fraction and John Lacy, all of the University of

emit their maximum energy near 3 fx. California, used a spectrometer working

Zirin completed an extensive study of in the far infrared with a tandem Fabry-
solar flares, comparing the optical emis- Perot-grating system having completely
sion in Ha and the A3835 band that he cooled optics so that sensitivity was
observed at Big Bear with data on ener- limited primarily by photon noise of
getic electrons reported by x-ray and radiation emitted in the path between
radio astronomers elsewhere. Among the the spectrometer and the object observed,
conclusions reached were the following: Observations were made of the 12.8-/X
(1) Hard x-ray emission appears only in line of Ne II in the galactic center and
impulsive flares having initial rise times in a number of planetary nebulae. The
of less than one minute. (2) No delay primary objective was to obtain velocity
was found between the start of hard and spatial distribution of the ionized
x-ray activity and the start of Ha emis- gas component of material in the galactic
sion. (3) The Ha intensity depends on center; Ne II provided one of the most
the soft x-ray flux; emission in the A.3835 convenient probes into this region. Spec-
band is observed when the soft x-ray tra were taken in 24 fields of view near
flux exceeds 5 X W phot cm^^ sec at the center of the Galaxy, each of 7"
the earth. In a number of instances, Ha diameter and with a spectral resolution
knots brightened simultaneously although corresponding to a velocity spread of
they were at considerable distances from about 40 km s~^. The spectra obtained
the source of activity. The exciting agent from the galactic center showed remark-
(hard electrons) must have traveled with able variations in intensity of the neon
velocities of at least 5000 km s~^. line and in its velocity distribution among

The 2.5-meter du Pont Telescope un- fields of view separated on the relatively

derwent fitting out during the year, with small scale of about 1/6 parsec. Most

124

CARNEGIE INSTITUTION

of the neon is redshifted by amounts
varying up to several hundred km s~^.
However, there are also substantial
amounts of blueshifted neon with equally
large Doppler shifts.

Also in the month of April, Babcock
and Brucato tested the wide-angle pho-
tographic capability of the du Pont Tele-
scope, using plates as large as 50 X 50
cm-. The diagonal of such a plate cor-
responds to 2.1° at a scale of 10.8" per
millimeter. Uniformly good star images
were recorded over the whole of each
plate in excellent seeing. One of the
large plates, of the cluster of galaxies in
Centaurus. is being used by Dressier and
Sandage in a study of the distribution
of galaxy types as a fimction of distance
from the center of the cluster; the pur-
pose is to understand the formation of
SO galaxies.

Auxiliary instruments scheduled for
installation, tests, and observations on
the du Pont Telescope before the end
of 1977 include the Cassegrain instru-
ment adaptor base, the Cassegrain spec-
trograph, and the infrared photometer.

An important part of the research con-
ducted with the facilities of the Hale
Observatories was accomplished by
astronomers from other institutions who
were invited to come as guest investi-
gators to conduct research projects ap-
proved by the Observatory Committee.
Some 58 astronomers from 38 institutions

were accommodated. Of particular in-
terest and value to astronomers of the
Hale Observatories were the opportuni-
ties to collaborate with guest investi-
gators who brought with them novel and
productive auxiliary instruments of types
not otherwise available. A partial list of
examples might include Dr. A. Boksen-
berg of the University College, London,
with his image-photon-counting system;
Dr. T. G. Phillips of Bell Telephone
with his heterodyne spectrometer; Dr.
Charles T. Townes of the University of
California with his helium-cooled spec-
trometer for the far infrared, and Drs.
J. Trauger and F. Roesler of the Uni-
versity of Wisconsin with their PEPSIOS
interferometer.

While basic support for the Hale Ob-
servatories is provided by the California
Institute of Technology (for the Palomar
Observatory) and by the Carnegie In-
stitution of Washington (for the Mount
Wilson Observatory and the Las Cam-
panas Observatory), a majority of staff
astronomers receive grant support for
their principal researches from federal
agencies: the National Science Founda-
tion, the National Aeronautics and Space
Administration, and the Office of Naval
Research. Thus many of the results listed
in this report depend importantly on
federal funds. Not less appreciated are
gifts from foundations, from industry,
and from individuals.

OBSERVING CONDITIONS

The 2.5-meter Hooker telescope on
Mount Wil.'^on was used for observations
on 244 complete nights and 121 partial
nights for a total of 2461 observing hours.
The telescope was not out of service for
repairs or engineering services requiring
more than two hours during any one
night. The 1.52-meter telescope was also
in regular use. At Mount Wilson the total
rainfall for the year was 750 mm and
the total snowfall was 1600 mm.

The 5-meter Hale Telescope at Palo-

mar Mountain was used for a total of
3063.6 hours, of which 2750.1 were night-
time hours, as shown in Table 1. The
difference represents twilight or daylight
time that was used for planetary or in-
frared observations not requiring a dark
sky. The Palomar 1.52-meter telescope
was in nearly constant use, while the
two Schmidt telescopes (1.2-meter and
46-cm) were in regular use except during
the light of the moon.
Total precipitation at Palomar was

HALE OBSERVATORIES

125

TABLE 1. Observations With the Hale Telescope

Zero

Total

Hours of

Complete

Partial

Observation

Hours

Nighttime

Month

Nights

Nights

Nights

Worked

Observing

July

27

4

236.8

263.8

August

29

1

1

265.7

265.7

September

14

8

8

160.0

160.0

October

22

7

2

274.0

264.0

November

23

3

4

284.5

274.2

December

23

5

3

318.7

280.8

January

18

3

10

265.6

214.6

February

24

3

1

341.5

278.0

March

20

6

6

251.5

229.5

April

24

5

1

305.1

219.7

May

10

10

11

135.4

117.6

June

24
. 258

3
37

3
60

224.8

209.2

Total

3063.6

2750.1

735 mm, with 960 mm of snow. The
maximum temperature was 34 °C, reached
in July 1976 and June 1977; minima of
— 8°C were reached in January and in
March.

Public visitors at the Palomar Ob-
servatory numbered 94,590 for the year.

At the Las Campanas Observatory in

Chile, the 2.5-meter du Pont telescope
underwent testing and commissioning.
The mirrors were collimated in March
1977, and the first scientific observations
were made in May. The 1-meter Swope
Telescope was used on 208 nights by 14
different observers. Snowfall for the year
amounted to 326 mm.

PHYSICS OF THE SUN

Mount Wilson Observations

Synoptic observations of the sun con-
tinue at Mount Wilson under the super-
vision of Howard. Between June 1, 1976,
and May 31, 1977, the following observa-
tions were obtained:

Direct photographs

279

Ha spectroheliograms,

9-meter focus

487

K2 spectroheliograms,

9-meter focus

482

Full-disk magnetograms

284

Integrated-light magnetic-field

measurements

231

Sunspot drawings

309

Sunspot magnetic-field

measurements

301

Solar observations were made on 317
days.

The daily magnetograms are published
in the monthly publication Solar Geo-
physical Data by the National Oceanic
and Atmospheric Administration. The
magnetic synoptic charts are published
in the Quarterly Bulletin on Solar Ac-
tivity of the International Astronomical
Union. Partial support for these obser-
vational programs comes from the Na-
tional Aeronautics and Space Administra-
tion, the National Science Foundation,
and the Office of Naval Research.

Since late in 1966, the digitized solar
magnetic and velocity data from the
Mount Wilson observations have been
preserved on magnetic tape. A pass
through this large amount of data to
study magnetic or velocity characteristics
over a long time interval is an expensive
and time-consuming enterprise. In order

126 CARNEGIE INSTITUTION

to facilitate a number of studies of long- some flux will be lost because some

term properties of the sun, a rereduction measurements will include field elements

of all the accumulated data has been of both signs, which will tend to cancel,

started. A part of the rereduction is the If during the approach to minimum

storage of a condensed data set for each (1974-1975) the polar fields became

day's observation. At the completion of more unipolar in each hemisphere, the

the project, expected during the next measured total flux could have increased

year, condensed data from the 11-year without any comparable increase in the

interval will be available on one mag- number of field lines because there was

netic tape. less cancellation of flux by opposite

polarity fields.

c 1 ^^ I- IT- ij Howard, in collaboration with Dr. Z.

Svestka of American Science and Engi-

Howard has estimated the true average neering, has studied the photospheric and
magnetic field strength at the poles of coronal magnetic-field configuration asso-
the sun. This quantity is of some im- ciated with a large complex of activity
portance because solar wind models de- in 1973. The Mount Wilson magneto-
pend critically on the true field strength graph measurements were used in con-
in the photosphere. Recent models of junction with the A.S. & E. Sky lab x-ray
coronal holes extended to the interplane- observations. The basic components
tary medium have led to inferred average (active regions) of the complex were
polar-field strengths of about 20 gauss, connected throughout its lifetime of seven
Since the magnetograph measures only solar rotations through systems of mag-
the line-of-sight component of the mag- netic field lines, including some across
netic field, it has at times been assumed the equator. The visibility of individual
that a strong polar field might exist but loops in these connections was greatly
escape detection. Howard's analysis, how- variable and typically shorter than one
ever, shows that the average polar fields day. Only the faint loops connecting
measured at Mount Wilson (in recent active regions with remnants of old fields
years, 1 to 2 gauss) cannot be low by can be seen in about the same shape for
more than a factor 5, and probably are many days. Brightenings of x-ray loops
not too low by more than a factor 2. were generally related to variations in
Within about 10° of each pole, the mea- the magnetic-field strength or configura-
surements are not reliable enough to use tion near one of the footpoints. The x-ray
in such an analysis. loops that interconnect active regions

Howard has found that, starting in generally connect to field elements in
mifl-1974, the total magnetic flux (F^ = the periphery of the regions — never to
i F+ I -|- I f- |) at latitudes poleward of sunspots. Some striking loop brightenings
40° in each hemisphere has increased by were associated with flares, but the loop
a factor of about 2.5. No comparable brightening and the flux occurrence
change was seen in the total flux equator- seemed to be two independent conse-
ward of 40°. Since the total flux poleward quences of a common triggering action:
of 40° latitude represents only about 5% the emergence of new magnetic flux,
of the total flux of the whole solar sur- Z. Svestka and C. V. Solodyna of
face, this is not a large effect. It may be A.S.&E., R. H. Levine of the Center for
that this is not a real flux increase on Astrophysics, and Howard have studied
the sun but a change in the distribution Skylab x-ray and Mount Wilson mag-
pattern of positive and negative field netic observations to determine whether
elements. The measured flux depends on or not open field lines may be found in
the aperture size because unless the aper- active regions. '^Opcn" field lines are
ture is the size of the field elements, those that extend into interplanetary

HALE OBSERVATORIES

127

space. It was found that some of the for the well-known 5-minute oscillation,

dark gaps seen between interconnecting no periods were found in the data. Upper

loops and interior loops of active regions limits of 1 ms-^ were established for

may be the loci of open fields, as has such periodic motions for the velocity

been predicted by global potential ex- signal, 0.02% for the (line-wing) in-

trapolation of photospheric magnetic tensity signal, and 0.03 gauss for the

fields. The field lines may be open only magnetic signal,
in a later stage of the active-region

development. j^-^ ^^^^ ^^^^^ Observatory

T o 7 T7 7 -i ET- 7j Durlug the year, the 65-cm vacuum

Larqe-bcale Velocity tields ^ , ^ ^^ ^ ^ ^• ^ ^ ^ ux

^ ^ telescope was aligned and light brought

The magnetogram reduction computer to the coude focus. The Zeiss universal
programs have been largely rewritten filter and associated systems were brought
by Boy den for the data rereduction under computer control. Diurnal drift of
project. As a part of the reanalysis, the the photoelectric guiders was found to
background velocities on the solar disk be due to differential flexure and was
were examined in detail. This involved removed by relocating the guiders. Auto-
subtracting the rotation, the limb red- matic camera controls have been added;
shift, and the effects of the earth's orbit six cameras may now be operated simul-
and rotation. After doing this, Howard taneously.

and Boyden discovered that an additional Observational programs centered on
effect remained — a low-amplitude, low- quiet-sun phenomena: spicules, solar ro-
latitude, symmetric east-west line shift tation, lifetime of the chromospheric net-
that varies linearly with the sine of the work, lifetime of quiet-sun magnetic
central meridian distance. A possible fields, nature of the helium chromosphere,
explanation is that the supergranular and mapping coronal holes,
motions (observed to be largely hori- Adams has acquired new data for in-
zontal) have a net upward component, vestigation of the rotation curve for
This would favor the motion toward the short-lived solar filaments. In an attempt
observer near the solar limb. As the re- to clarify the geometry of the magnetic
reduction proceeds, more data will be- field involved in the solar-cycle field-
come available with which to examine reversal, Adams has developed a model
this and other possible explanations. in which surface reconnection of flux

from decaying active regions provides

Solar Puhatiom ^^^^ necessary link in producing the

reversal.

A series of observations was made in Hurford has continued his work with

August 1976 at the 150-foot Tower Tele- the NASA-constituted Hard X-Ray

scope in an attempt to detect global Imaging Facility Definition Team on the

pulsations of the sun. Such pulsations design of an imaging instrument. This

with an amplitude of 5 ms-^ and a period Shuttle-borne instrument will provide 4"-

of 160 min have been reported by ob- resolution images of solar flares and

servers at the Crimean Astrophysical cosmic sources between 2 and 80 kev in

Observatory. Also, Hill and his associates order to better understand the behavior

at the University of Arizona have made of nonthermal particles. During the past

solar diameter measurements that indi- year he has developed an improved x-ray

cate low-mode pulsations with periods in collimator configuration that permits

the range of 50 min. The Mount Wilson high-resolution wide-field imaging,

observations w^re examined for pulsa- Barry J. LaBonte, a graduate student,

tions between 4 min and 180 min. Except worked on measurement of the helium

128

CARNEGIE INSTITUTION

D3 line profile with a birefringent filter.
For plages on the sun's disk, he found
the best-fit Gaussian has a 1/e width of
0.4 A. with negligible instrumental con-
tribution. The D3 opacity is produced in
regions with thermal line width — 0.1
A; the nuich larger observed width indi-
cates large nonthermal motions in the
chromosphere.

Microphotometry of calcium iv-line
photographs in the regions of polar
coronal holes has been carried out by
K. A. ^larsh. who foimd that the chro-
mospheric network exterior to a hole has
a slightly broader intensity distribution
than that inside the hole, a fact that can
be attributed to a greater number of
bright network elements outside the hole,
^larsh has also made a study of high-
resolution filtergrams of the solar limb
in D3 and ofT-band Ha in order to in-
vestigate the spatial structure of the D3
chromosphere. It was found that spicules
provide the major contribution to the
peak D3 intensity, with the remainder
of the emission coming from a semi-
homogeneous background component at
low height. The observations can be un-
derstood on the basis of the photoioni-
zation model, whereby it was found that
helium is only slightly ionized at the
height of the D3 peak, and that spicules
are at least three times denser than their
surroundings at this height. In coronal
holes, the D3 emission is confined to iso-
lated patches, but these patches have the
same basic character as the spicule
bushes that comprise the normal chro-
mosphere.

Macro spicules, X-Ray Bright-Point
Flares, and Spicules

Two solar phenomena that were newly
ob.served from Skylab are extreme ultra-
violet macrospicules and flares in x-ray
bright points. Moore and F. Tang, in
collaboration with J. D. Bohlin at the
Naval Research Laboratory and L.
Golub at American Science and Engi-
neering, have completed an extensive

comparison of Ha observations from Big
Bear Solar Observatory with EUV and
x-ray observations from Skylab; they
find that macrospicules and x-ray bright-
point flares are closely related.

Moore and Tang found that time-lapse
Ha filtergram movies of the solar limb in
quiet regions show small surgelike erup-
tions that are quite similar to EUV
macrospicules in size, shape, motion, and
duration. They therefore named these
transient Ha features Ha macrospicules.
From the similarity of Ha macrospicules
and EUV macrospicules and from com-
parison of simultaneous Ha and He II
304 A observations, Moore et al. conclude
that Ha macrospicules are EUV macro-
spicules viewed in Ha, although most
EUV macrospicules are too faint in Ha
to appear on Ha filtergrams of normal
exposure.

From comparison of simultaneous x-
ray and Ha observations of flares in
x-ray bright points situated on the limb,
Moore et al. find that flares in x-ray
bright points often produce Ha macro-
spicules. This suggests that all macro-
spicules may be produced by similar,
but usually weaker, microflares and that
spicules may be generated by still smaller
microflares.

X-Ray Spectra of Flares

Moore has analyzed the soft x-ray
spectrum of 11 flares observed simul-
taneously from the SOLRAD-9 satellite
(A < 20 A) and from the OSO-7 satellite
(A < 3 A). He found that both spectral
ranges gave the same temperature and
emission measure for the emitting flare
plasma to within the accuracy of the
data; i.e., the A < 20 A flux and the
A < 3 A flux are both emitted from ap-
proximately the same plasma. The de-
rived temperature of this plasma at flare
maximum is invariably in the range 1-2
X 10^ K. Solar plasma in this tempera-
ture range emits about 80% of its radi-
ation shortward of 20 A. Therefore, the
comparison of SOLRAD-9 and OSO-7

HALE OBSERVATORIES

129

thermal x-ray flux observations shows
that the A < 20 A soft x-ray flux is a
good measure of the total radiative loss
rate from the 7" > 10^ K flare plasma,
while the A > 20 A XUV flux is a good
measure of the radiative output from all
lower temperature (T < 10"^ K) com-
ponents of the flare.

Physics of Flares

Zirin completed an extensive study of
solar flares, with the object of under-
standing the temporal, spatial, and ener-
getic relations between energetic elec-
trons (observed by x-ray and radio) and
optical emission in Ha and the A 3835
band. The following conclusions were
reached:

1. For about 50 flares with good x-ray
and spatial data, an average lag of ± 3
sec was found between Ha start and
hard x-ray (HXR) start, as well as be-
tween Ha peak and HXR peak. If the
soft x-ray (SXR) peak lags the HXR
peak, the Ha brightness will continue at
the peak value, decaying with the SXR
flux.

2. The Ha intensity depends on the
SXR flux. Emission in the 3835 band is
observed when the 5.1-6.6 keV SXR flux
exceeds 5 X 10^ phot cm-^ sec at the
earth. HXR emission appears only in
impulsive flares with initial rise time of
less than one minute.

3. Multiple HXR spikes are connected
with complex flares by a series of sepa-
rate optical phenomena; each spike
appears to be connected to a distinct
happening rather than some oscillatory
process. In some cases, however, homolo-
gous flares will occur within hours at the
same place with the same Ha and x-ray
properties.

4. In several cases. Ha knots sepa-
rated by considerable distances appeared
simultaneously. The exciting agent must
have traveled at least 5000 km s~^ and
(since an HXR spike was simultaneous)
the cause must have been hard electrons.

5. In all cases where flares near the
limb were observed, an elevated Ha
source was seen. Normally this was a
preexisting mound or arch of cooler ma-
terial suddenly excited by the onslaught
of hard electrons. Calculations show that
the cool material will be heated within
10-100 sec to SXR source temperatures,
and about 10% of the electron energy
will come out in Ha, the rest going to
thermal energy.

Solar Radio Astronomy

The solar interferometer equipment at
Owens Valley Radio Observatory has
become operational, and Hurford, Marsh,
and Zirin have been carrying out a pro-
gram of solar radio observations at 2.8
cm with two objectives: (1) to study the
emission and polarization of flares and
active regions, and (2) to study the
small-scale structure of the quiet sun
to determine the upper chromospheric
temperature distribution. There is little
to say about the active sun because there
are few flares in this quiet time. How^-
ever, one flare wath a flux of 4 X 10~^
solar flux units (400 Jy) was observed.
Optical changes in the observed flare cor-
respond to phase shifts in the positional
fringes, and analysis is under way to
determine the relative position of the
optical and radio source.

Quiet-sun observations have shed some
light on the puzzling quiet-sun fringes
observed years ago by Lang and Zirin.
By observations at several baselines, it
has been found that (1) the fringes are
uniform over the sun; (2) the fringe
amplitude increases as resolution de-
creases; and (3) the sources are un-
polarized and there is no correlation be-
tween different baselines.

A model has been proposed in which
the sources are a random superposition
of components of the chromospheric net-
work. These thermal sources are about
10^ °K hotter than the background. This
model fits fringes observed at 8.6 cm
and 11 cm as well.

130

CARNEGIE INSTITUTION

SOLAR SYSTEM

lo

With the 150-mm PEPSIOS spectrom-
eter of the I'niversity of Wisconsin in-
stalled at the coude laboratory of the
5-nieter telescope. Miinch. in collabora-
tion with Drs. Fred Roesler and John
Trauger of the University of Wisconsin,
has studied the [S II] AA 6716-6730 lines
emitted in- the plasma contained in the
Jupiter magnetosphere. During the nights
of October 7-11, 1976, 70 line scans were
obtained with a resolving power of
130.000 and a circular field of view^ of
30" diameter. An example of the quality
of the tracings is shown in Fig. 1, which
shows the raw data (step graphs) super-
posed by Gaussian forms fitted by least
squares. From the study of the material
available, the following conclusions have
been drawn:

1. The S+ ions are corotating with the

Jupiter magnetic field. When we adopt
the rest wavelengths determined by meas-
urements in spectra of emission nebulae,
a small systematic difference appears
between the observed and expected wave-
lengths in the corotating frame. We
ascribe it to inaccuracy of the rest wave-
lengths, which should be increased by
0.05 A for both lines.

2. If the distance r from Jupiter is ex-
pressed in terms of a(Io) , the semi-major
axis of lo's orbit, then the maximum at
r := 0.8 a(Io). No emission was detected
at r = fl(Io), although our field of view
had a diameter of 0.2 a(Io). The surface
brightness in the three tracings obtained
at r = 0.5 a(Io) appears to be only half
of its maximum value at r = 0.8 a(Io).

3. The S+ emission is strongly con-
centrated toward the plane of lo's orbit.
No emission was observed when the field
of view had a magnetic latitude of ±20°.

5x10"^

I xiO'

5XI0-

5(6717)
S(673l)

1.0 0.9 08

Np~N(S*)

0-

10'

N3(cm^)

Fig. 1. Repre.sentative scans of tho X6716.47 and \6730.85 lines of singly ionized sulfur in
Jupiter'.s corotating pla.'^ma, as obtained at 7.5'' UT on October 11, 1976. The spectra
characterize a 30" field of view centered at 0.8 lo orbital radii west of Jui)iter and within
2'^ of Jupiter's magnetic equator at a time when lo was near eastern elongation. Nearly
equal 45 Rayleigh surface brightnesses are obtained for these two lines.

HALE OBSERVATORIES

131

4. The S+ emission rate appears to
depend on the relative geometry of the
Jupiter magnetic field, the orbital longi-
tude of lo, and the line of sight. The
dependence, however, cannot be made
precise on the basis of the available ma-
terial because the number of tracings is
insufficient to extricate the various time
factors involved in the intensity vari-
ations of the lines.

5. The width of the S+ lines is between
7.0 and 7.5 km s~^. After allowing for
instrumental broadening, the intrinsic
width of the lines, if assigned to pure
thermal motions, corresponds to a tem-
perature of 2,000° K.

6. The doublet intensity ratio [A6718] /
[A6730] has the value of 0.90 ± 0.10,
which constrains the values of the elec-
tron density Ne to the range indicated in
Fig. 2. The given uncertainty for the
value of the doublet ratio is not due to
noise in the tracings but rather to the
time variability of the line intensities.
The constraint Ne — A^(S + ) on the ion
density A^(S+), together with the values

CO

z

LxJ

J I I \ \ L

-0.2 0.0 +0.2
WAVELENGTH -Xo( A)

Fig. 2. Electron density and temperature in
Jupiter's co-rotating plasma at a distance 0.8 lo
orbital radii from Jupiter. The range of allowed
values is constrained by the observed ratio of
[S II] line strengths, the observed line widths
(Tg < 30,000°K), and charge neutrahty of the
plasma. The data are representative for obser-
vations during the period S-S*" UT on October
11, 1976.

of T obtained from the line widths, de-
fines the domain indicated in Fig. 2.
MiJnch concludes that the heavy ion
abundances approximate those expected
for a meteoritic composition; namely, in
proportions Na:Mg:Si:S:Fe = 0.04:1.0:
1.0:0.6:1.0. In particular, the H-, He-,
and CNO-group ions cannot be present
in any proportion remotely approaching
that expected for a Jupiterlike compo-
sition.

It is believed that this result has far-
reaching conceptual implications on the
origin of the Jupiter plasmasphere, as it
means that the confined plasma torus
seen in S+ does not ''communicate" with
Jupiter, nor does it bear the signature
of the solar wind. The evidence thus is
overwhelmingly in favor of a satellite or
meteoritic origin for the thermal plasma
observed in the Jupiter magnetosphere.
But this conclusion raises a difficult
problem: Atoms leaving lo, upon col-
lision ionization, produce ions that have
a velocity of 57 km s~^ with respect to
the magnetic field on which they are
trapped and therefore will gyrate wdth
this velocity around their guiding centers.
The free electrons will thermalize by
elastic scattering in a few minutes to an
assembly characterized by kinetic tem-
perature of about 10^ °K. If the entire
plasma could relax into equilibrium by
particle-to-particle collisions, it would
reach a temperature half as great, after
undergoing ionization and excitation, but
the time required would be of the order
of eight months. It is most likely the
collective plasma instabilities will set in
within a much shorter time scale, which
somehow will dissipate the high internal
energy. The energy per ion-electron pair
created on ionization of an atom leaving
lo, in the corotating frame, is 7 X 10~^°
erg, and, for a flux of Na atoms lea\'ing
the lo surface of 10' atoms cm~- s~^,
the total energy being injected into the
plasma is 2 X 10^" ergs s~^. This energy
has to be dissipated before the plasma
settles down to a temperature of 20.000°K,
as observed, but the process through

132

CARNEGIE INSTITUTION

which the dissipation takes place is not scope was begun by Kowal in December
yet clear. I'ndoubtedly some ''dumping" 1976. The main purpose of this program
into the Jupiter ionosphere takes place is to study the distribution of very dis-
hy diffusion along the magnetic lines of tant comets. In addition to comets, any
force of Coulomb-scattered ions and new objects nearer than Mars or more
electrons, but radiative losses from the distant than Jupiter will be followed,
hot plasma (5CX).000'^K) will also occur. Thus far, two ''lost" objects have been
If radiation is the predominant loss recovered and two new objects have been
mechanism, it would balance the energy found. The "classical" Apollo-type aster-
input for a hot ion density of 100 cm~^. oid Adonis was recovered in February
The fact that the S+ emission originates 1977. This object had not been seen
in a torus centered not at r = a(Io), since 1936. Periodic Comet Taylor (1916
where the neutral atoms leaving lo be- I) was found on plates taken in Decem-
come ionized, but at r = 0.8 a(Io) sug- ber 1976. This comet had not been seen
gests the existence of the hot plasma at since 1916 and was thought to be hope-
r ^ a(Io). This hot plasma could have lessly lost. It split into two pieces in
been detected by the Plasma Analyzer 1916. Only component "B" is now visible.
Experiment of Pioneer 10. The radiation A new Apollo-type asteroid, designated
detected in the vicinity of lo's orbit by 1977 HB, was found by Kowal in April
the short-wavelength channel of the 1977. The orbit has a semimajor axis
Pioneer 10 UV photometer could also be of 1.08 A.U., making the asteroid a can-
the line emission of the hypothetical didate for a possible space probe. In
coronal plasma ^Ig X AA625-610, Si XII terms of energy requirements, 1977 HB
AA520-50. Mg IX A368, etc. Further is probably the easiest object to reach,

theoretical study of the plasma physics
problems involved will be required to
provide a basis for these tentative con-
clusions.

other than the Moon.

The first new comet of the program
was found in April 1977. Comet Kowal
(1977 f) has a period of 15 years and a
perihelion distance of 4.65 A.U.

This project is under the general di-
rection of Mlinch and is funded by Na-
A systematic survey of the ecliptic tional Aeronautics and Space Adminis-
region with the 1.2-meter Schmidt tele- tration Grant NGL 05-002-140.

Comets and Asteroids

STELLAR SPECTROSCOPY

White Dwarfs

The use of the SIT-Vidicon Cassegrain
spectrograph and the multichannel spec-
trophotometer has resulted in discovery
of over 50 additional degenerate stars
by Greenstein and collaborators. At-
tempts are being made to extend the
search for these objects fainter than
magnitude 18. Observations with the
SIT. in particular, have shown that
higher resolution (about 5 A) permits
detection of weak metallic lines and,

particularly interesting, very weak hy-
drogen lines in cool degenerate stars. The
hydrogen lines persist down to tempera-
tures well below 7000°, even though one
expects that, because of convection, the
mixing of the outermost hydrogen layers
and the deeper layers of more normal
composition should be bringing metal to
the surface. Thus continued evidence
exists for stratification in the composition
of the degenerate stars and the presence
of relatively sharp but weak hydrogen
lines in quite cool objects which are

HALE OBSERVATORIES 133

almost certainly dominated by helium, jects are clearly among the lowest lu-

The number of intrinsically faint red minosity stars known; some are of large

degenerate stars has been substantially space velocity as well as: low luminosity,

increased so that there are now approxi- which is rare. An example is LP 16-36

mately 50 known red degenerates of with a visual absolute magnitude My ^=

temperature 7000° or lower. A few have 16.5 and a space motion of 200 km s-^

quite strong metallic lines. One, GD 401, In addition, in collaboration with George

has H and K of 175 A equivalent width. Gatewood of the University of Pitts-

An especially interesting object, LP93- burgh, who plans parallax studies of the

21, is clearly underluminous for its colors, very faint stars of the LP survey, some

presumably because its atmosphere has intrinsically very faint M dwarfs have

extremely strong molecular carbon bands, been found, including two near Mv +

The SIT permits resolution of the vibra- 17.9, among the faintest stars known,

tional structure and suggests equilibra- Such stars can be found only if close to

tion between vibrational, rotational, and the sun; any appreciable number will

local temperatures. From colors, LP93- substantially increase our knowledge of

21, which has a proper motion of 1.77'' the frequency of stars of the smallest

per year, would be given a luminosity mass,
such that its tangential space motion
would be over 1000 km s~^. In fact, the

star must be intrinsically fainter and B-Tvvp S^nrs
closer than the colors would indicate, but

even with a maximum reasonable cor- Preston and Sidney Wolff of the Uni-
rection, which reduces its radius to one- versity of Hawaii have completed a study
half the mean of the average degenerate of the rotations and the incidence of
star, the space motion is nearly 600 km MgMn stars in a large color-limited
s~i with a retrograde and eccentric sample of B-type stars. Projected rota-
galactic orbit. A pair of red degenerate tional velocities with a resolution of 10
stars, LP380-5/6, have been studied in km s~i were obtained from Palomar
cooperation with W. F. van Altena of coude spectrograms. HgMn stars were
Yale University, who has obtained a discovered by examination of these spec-
parallax for those objects with the Kitt trograms and, additionally, by a moder-
Peak National Observatory 4-meter re- ate dispersion (50 A mm"~^) survey at
flector. The two stars are among the Mauna Kea of the strong ultraviolet
intrinsically faintest known, with a visual lines of Mn II. For the latter survey,
luminosity approximately 10~^ that of discovery probability for Mn stars is
the sun. virtually independent of rotation for v

sin i < 100 km s~^. Inversion of the

T 17 ^ I ±-L nr ' o i^sini distribution for the whole sample

Lower hind 0] the Main Sequence , -r , .^ ^. ^ ^ - ■ a ^ ,^

by Lucy s iterative technique yields the

In the survey of proper-motion stars rotational velocity distribution for late

from the Lowell Observatory catalogs B-type stars. The distribution rises from

of Giclas and collaborators, the apparent a small (indeterminate) value at zero

magnitude limit of 16.5 results in a limi- velocity to a maximum near 50 km s""^

tation on the lowest luminosity stars and then declines at higher velocities.

likely to be so observed. Greenstein has The frequency of Hg]\In stars decreases

been able to locate and observe numer- monotonically from 60% of the sample

ous fainter red M 'stars discovered by for ?; < 10 km s"^ to zero at v ^ 100

W. J. Luyten of the University of Min- km s~^. Thus slow rotation appears to

nesota on the Luyten-Palomar Schmidt be a necessary but not sufficient condi-

(LP) proper-motion survey. These ob- tion for the occurrence of the abundance

134

CARNEGIE INSTITUTION

anomalies of the HgMn stars. A second
parameter, vol to bo identified, must be
involved.

Red Giants in Globular Clusters

Zinn and John Xorris of Mount Stromlo
Observatory have completed a spectro-
scopic search of the globular clusters
M13. M15. M92. and NGC 6397 for red
giant stiirs that have unusually strong
or weak CH and CN bands. In a sepa-
rate study. Zinn has completed a similar
survey of ^15.

These surveys have revealed that
roughly 80*^ of the asymptotic branch
stars in each of these clusters have un-
usually weak G-bands (the CH band at
4310 A I. which confirms the results of
Zinn's earlier, less extensive survey of
M92. In addition, a milder G-band vari-
ation has been detected among the stars
on the giant branch at roughly the mag-
nitude of the horizontal branch. To see
if these G-band variations are the result
of a variation in metal abundance, Norris
anrl Zinn obtained intermediate-band

photometry of selected stars in M13,
M15, and M92 with the 5-meter Hale
telescope. These observations did not de-
tect a dift'erence in metal-line blocking
among stars of widely different G-band
strengths, which suggests that the weak-
G-band effect is produced by stellar evo-
lution and not by primordial inhomo-
goneities within the clusters. Norris and
Zinn believe the weak-G-band effect is
caused by a mixing process that mixes
the outer parts of a star with its interior
regions that have been depleted of car-
bon by the CN cycle. To explain the G-
band variations seen on both the giant
branch and the asymptotic branch, it
seems necessary to hypothesize that mix-
ing occurs at least twice during a star's
giant-branch evolution. Although none
of the theoretical calculations of stellar
evolution predict the weak-G-band effect,
the observational data seem to suggest
that mixing occurs first on the lower
giant branch when a star's surface-
convective zone attains its greatest in-
ward extent and again at the tip of the
giant branch during the helium-core flash.

STELLAR CHROMOSPHERES

Instrument for the Study of
Stellar Chromospheres

In June of 1975, Vaughan, Preston,
and 0. C. Wilson as principal investiga-
tors received a grant from the National
Aeronautics and Space Administration
for construction of an instrument de-
signed specifically to measure stellar
chromospheric H- and K-line fluxes. The
instrument was successfully put into
operation in April 1977 at the Cassegrain
focus of the 1 .6-meter telescope on Mount
Wil.<on. A single photomultiplier is used
to measure sequentially, at a chopping
frequency of about 30 Hz, the light flux
in each of two bands of 1 A width at the
wavelengths of Ca II H and K and in
each of two bands of 20 A width in the

continuum longward and shortward of
H and K. This is accomplished by a
dedicated third-order grating spectrom-
eter of modified flat-field Ebert config-
uration and an associated microprocessor-
controlled data system. By offsetting the
position of the entrance slit of the spec-
trometer, whose dispersion is 4.35 A
mm~^ and whose zero-point wavelength
is known with reference to a hollow
cathode source emitting Ca II H and K,
the observer can introduce a precise
Doppler-shift correction to the wave-
length of each observed star. Simplicity,
rugged construction, and the relatively
specialized nature of the instrument are
features intended to promote efficiency
and long-term stability in measurements
of stellar chromospheric activity.

HALE OBSERVATORIES

135

The measured data from observations
made at the 1.5-mctcr telescope, in col-
laboration with Wilson at the 2.5-meter
telescope, were transferred to the scale
of mean H-K fluxes developed for main-
sequence stars over the past decade by
Wilson with a digital system at the 2.5-
meter coude spectrograph. Extension of
Wilson's long-term study of normal
main-sequence chromospheric variability
to the new instrument is thus assured.

Flare Stars

A faint dM4e star discovered by Bond
on the University of Michigan objective-
prism survey plates flared by nearly 8
magnitudes. Greenstein has used the
multichannel spectrophotometer and the
SIT-Vidicon to study the outside flare.
He finds evidence of substantial altera-
tion of the energy distribution in the

ultraviolet. The Balmer emission lines
should be accompanied by a Balmer re-
combination continuum, which was, in
fact, observed as a flattening out of the
stellar flux in the ultraviolet and possibly
a veiling of the absorption lines. A model
for this star's chromosphere suggests
quite high density, a temperature of the
order of 20,000°K, and probably con-
tinuing activity between major flares.

In the calibration of the lower main
sequence by the multichannel spectro-
photometer for parallax stars, it has been
found that those known to be strong
emission-line or flare stars give the
largest dispersion in a luminosity-color
diagram. It is possible that the overlying
nonthermal continuum affects the ob-
served stellar flux to a wavelength as long
as 4500 A. This is consistent with older
observations in which the broad-band
colors U — B, B — V showed abnormal
scatter among late M dwarfs.

MILLIMETER-BAND PHOTOMETRY

One-Millimeter Photometry of
Extragalactic Objects

For the past three years a program of
photometry of extragalactic objects at A
=1 1 mm has been conducted in twilight
hours from the prime focus of the 5-
meter Hale telescope by members of the
infrared group. This is the most syste-
matic study of extragalactic objects yet
carried out at this wavelength and the
first which permits a search for 1-mm
variability. The results of the first three
years of this work have been prepared
for publication by graduate student J.
Elias.

A total of 23 extragalactic objects have
been observed at 1 mm, and 9 have been
detected. Those detected include two, the
galaxies NGC 253 and NGC 1068, from
which the 1-mm emission appears to be
thermal radiation from dust. The other

seven sources detected appear to be emit-
ting nonthermal radiation at 1 mm. These
objects are 3C84, 30111, 3C120, 3C273,
3C274(M87), 3C279, and BL Lacertae.

Thermal Sources

The 1-mm fluxes for the thermal
sources lie on an extrapolation of the
strong far-infrared emission observed
from these galaxies and attributed to
thermal emission from dust. The 1-mm
data have been used to produce estimates
of the mass of dust and gas in the central
regions of the galaxies. The mass of gas
is in good agreement with that inde-
pendently estimated by radio astronomers
who have observed CO emission from
the same sources. This agreement appears
to support the assumption that the CO
emission from these galaxies is not hea^^ly
saturated.

136

CARNEGIE INSTITUTION

2
1.5

I -

0.5-

-3

05-

o

05
1.5

I-

WAVELENGTH
lOcm Icm Imm 100/J.m lOcm Icm Imm 100/j.m

_

I 1

I

I

O

3C84

-

I-

-

•

3CMI

• a

a

1

3CI20 _

• • □

a

^

-

• •

D

D

3C273

-

1

1

o

1

\ ,

1 1

1 1

M87 (3C274)

2

•

15

•

A

"

1

—

▲

{

0.5

-

1

• • □

3C279

D

i

05

-

-

1

-

BL Lac

0.5

• • °

D

i

1 1

1

1 1

13
LOG 1/

9
HZ

10

Fig. 3. Spectra of nonthermal sources. The open circles are the 1-mm measurements made
on the 5-meter telescope. The plotted jQuxes at each wavelength are those measured in January-
Februarv 1976.

Xonthermal Sources

The energy distributions of the non-
thermal sources are shown in Fig. 3; the
1-mm flux measurement from the 5-meter
telescope is given by the open circle. Note
that for each source the 1-mm flux is
consistent with what would be expected
from a smooth extrapolation of the longer
wavelength radio spectrum. Fig. 4 shows
for several objects the measured 1-mm
fluxes against time, and for comparison
the 2.1 -cm radio fluxes as a function of
time. Four of the sources, 3C84, 3C120,
3C273, and BL Lac show formal evidence
for variability. The evidence is most sig-
nificant statistically for BL Lac; note
also that this source varies by more than
a factor of 2 in 1-mm flux and that the
1-mm and 2.1-cm flux variations are well
correlated.

The principal result of the work on
nonthermal sources at 1 mm is that there
is no evidence in these objects for the
common occurrence of very compact
components that radiate chiefly at 1 mm
and shortward. This conclusion follows
from the good agreement of the 1-mm
flux with the extrapolated radio spectrum
(Fig. 4). It is supported by the fact
that no strong variations are seen at 1
mm that are not correlated with longer
wavelength variations. A second major
result is that although some variations
occur, strong 1-mm variability is not
invariably present in compact, nonther-
mal extragalactic objects. For example,
none of 18 separate measurements of
3C273 over a period of three years shows
convincing evidence for a value of the
flux that deviates by more than 50% from
the mean value for this time interval.

HALE OBSERVATORIES

~1

137

—;>

X

ZD

60
40
20

20
10

60
40
20

20-

10

• ••••••••

« •• • •

U

3C84

I \

3CI20

• •

* • • • * •

• • • • •

3C273

»• •

• •••• • • .

M 5^ J .

3C279

• •

'd^

§

• • •

•* - 1 •••'••••.^*^<

^?^*

§

BL Loc

1975.0

1976.0
DATE

1977.0

Fig. 4. 1-mm fluxes plotted against time for selected objects (open circles). Also shown are
2.1-cm observations (filled circles) kindly provided by Dr. H. Aller of the University of
Michigan.

Observing Procedures

The absolute flux measurements de-
scribed above were possible only because
considerable effort was expended in es-
tablishing a calibration system for these
observations and verifying its reliability.
For observations at 1 mm, this presents
special difficulties because of the large
fractional bandwidth (A/ A A ^ 1) used
for the observations and the large and
variable extinction due to atmospheric

water vapor. However, it is believed that
the uncertainties in the flux determina-
tions due to possible systematic calibra-
tion errors are no greater than 20% for
1-mm observations on the 5-meter tele-
scope. The sensitivity of the detector
system now in use permits a lo- limiting
flux of about 1 Jy (=10-26 W m-^
Hz-^) in one hour of observing under
typical observing conditions, and most
of the extragalactic objects detected have
fluxes < 5 Jy at 1 mm. It is hoped that

13S

CARNEGIE INSTITUTION

the sensitivity will be improved dramati-
eally in the near future to permit l-nim
observations of a niueh more extensive
set oi extragalaetie objeets.

Submillimctcr Heterodyne Spectroscopy

A new proi^ram of heterodyne spec-
troscopy at submillimeter wavelengths
lias been begun at the 5-meter Hale tele-
scope in a collaboration between Drs.
T. Phillips and P. Hiiggins of the Bell
Telephone Laboratories and Werner and
Xeugebauer. It has resulted in the first
detection in an astronomical source of
the J = 3^.1 = 2 CO line at 345 GHz.
The receiver, based on an InSb hot-
electron bolometer mixer, is mounted at
the prime focus of the telescope. Both
twilight and nighttime observations have
been made with this system. The 3^2
CO line has been mapped over the central

few arcminiites of the Orion Molecular
Cloud. Broad ('-^40 km s~M velocity
wings are seen on the line profile at the
center of the soiu'ce; these wings are
much more pronoimced than are the
similar wrings seen in the 2->l and l->0
lines, suggesting that the emission comes
from a small, optically thin cloud of CO.
The good beam profile and precise point-
ing obtainable at this frequency with the
5-meter telescope permitted us to demon-
strate that the high-velocity CO emis-
sion is confined to the central 30'' of the
source and that the emission peak is
within 10'' of the Becklin-Neugebauer
and Kleinmann-Low infrared sources.

In addition to the Orion observations,
3-^2 CO line emission has been detected
from Ml 7 and from the shell of molecular
gas around the carbon star IRC4- 10216.
Additionally, the J = 3-^2 emission
lines of HCN and HCO+ at 270 GHz
have been detected in the Orion region.

GLOBULAR CLUSTERS

Gunn and Dr. R. F. Griffin of Cam-
bridge Observatories, England, have com-
pleted their observations and analysis of
M3. using 108 high-accuracy radial
velocities of individual giants in the
cluster obtained with their cross-correla-
tion spectrometer. The cluster has been
modeled with a multicomponent King-
type distribution function w^ith anisot-
ropy a la Michie. It is clear that the
velocity dispersion falls off much more
rapidly with radius than predicted by
i.sotropic models, and the best fits have
been obtained with models whose anisot-
ropy radius is about 15 core radii, at
which point the two-body relaxation
time to deflections is roughly the Hubble
time. The mass-to-light ratio is higher
than generally suspected — about 3 in
visual solar imits for the cluster as a
whole. This high ^}/L must reside in
part in dark remnants of mass compara-
ble to the turnoff stars and giants, though

much of it must be in an extended lower
main sequence wdth a quite steep mass-
function. One important result is that
there is very little mass in remnants with
individual masses much in excess of the
turnoff mass, and there is some difficulty
in accommodating even 1.2 WIq white
dwarfs and neutron stars in their ex-
pected numbers. The velocity dispersion
"hole" in the center persists but is of
marginal statistical significance.

Infrared Studies of Red Giants in
Globular Clusters

Persson, Dr. J. Cohen of Kitt Peak
National Observatory, and Dr. J. Frogel
of Cerro Tololo Interamerican Observa-
tory completed an infrared study of a
selection of red giant stars in the globular
clusters M92, M3, and M13, and in M67.
The observational data, which consist of

HALE OBSERVATORIES

139

narrow-band CO absorption indices and
broad-band colors, were obtained with an
InSb system on the 5-meter telescope.
The broad-band colors were compared
with a new set of model-atmosphere
fluxes computed for metal-poor stars. An
important result of this comparison is
that V — K colors (i.e., V — [2.2 /x])
give a measure of stellar effective tem-
perature that is essentially independent
of metal abundance or surface gravity.
(This is because the H~ opacity is domi-
nant at both wavelengths.) The accuracy
attainable by this technique is currently
±100°K in Teff, and can be improved.
The stellar energy distributions were in-
tegrated to derive bolometric magnitudes
directly from the data, and thus a theo-
retical HR diagram (log L/L^ versus
log Teff) was constructed and compared
to the predictions of published models.
The shape and location of the theoretical
tracks, the dependence on metal/hydro-

gen ratio, and the location of the red
giant tip are in good agreement with the
observational results.

Because of the importance of Rayleigh
scattering in the metal-poor atmospheres,
B — V becomes sensitive to surface
gravity and can be used to derive crude
values for log g. With derived values of
T'eff and L/Lq, one can then derive
masses from the photometric data. These
masses are consistent with turnoff masses.
Refinement of the models and data could
lead to substantial improvement in these
determinations.

The CO absorption data show clearly
that the clusters are ordered in the se-
quence M92, M13, M3 (toward increas-
ing CO). This agrees with the ordering
in the HR diagram. Metal-abundance
determinations, however, give the order-
ing as M92, M3, M13. The conclusion
is that CNO abundance variations can
be different from those in Fe.

INTERSTELLAR MATTER AND GASEOUS NEBULAE

The infrared source in the so-called
Red Rectangle connected with HD 44179
was observed with the multichannel
spectrophotometer by Greenstein and
Oke. Unlike all other reflection nebulae
observed, the nebula is appreciably
redder than the star. The other peculiar
feature in the Red Rectangle is the
presence of a very large increase in flux
centered on A6500 and about 1500 A wide.
Although this is near Ha, it is almost
certainly not hydrogen but a feature of
the reflectivity of the dust near that star.
The usually accepted model, a torus in
the line of sight that obscures light di-
rectly from the star but lets it leak out
to illuminate dust in the plane of the
sky, has been analyzed in some detail
with the radiative-transfer methods de-
veloped by Chandrasekhar for the illumi-
nation of atmospheres by an external
source. Using the so-called exact solu-
tions to this theoretical problem, Green-

stein and Oke find that the high flux
scattered in the red requires a large in-
crease in the albedo of the dust. Since
so many geometrical factors are un-
known, for example, the distance from
the star to the nearest point of the nebula
we can see, a most favorable case can
be constructed in which the albedo is in
general near 0.4 but rises to 0.7 in the
red. If on the other hand, the illumina-
tion properties of the dust are less favor-
able, that is, if the dust is at a greater
distance from the star, a larger albedo
is required. Quite modest changes in the
assumed model result in an albedo in
the red which exceeds unity and a gen-
eral albedo in the rest of the spectral
range 0.6 to 0.7. Four other observed
reflection nebulae with stars showing
infrared excesses fail to display this
peculiar high red reflectivity. The min-
erals in the dust cannot be identified
from the single observation of high re-

140

CARNEGIE INSTITUTION

flectivity in the rod; the redness would
be that of a brilliant red paint but with
higher reflectivity than most known ma-
terials. One interesting but speculative
{.x)ssibility. drawn from the literature on
the reflectivities of minerals, is that some
com{^ounds containing manganese have a
peculiarly strong red fluorescence.

Planctanj Xcbulae

De Bruyn. in collaboration with G.
Wouterloot and Dr. J. H. Oort of Leiden
Observatory, used the 1.2-meter Schmidt
to take plates through an Ha-interference
filter of 80 A bandpass of several fields
close to the galactic center. Matching
short-exposure red continuum plates
were also obtained. The plates will be
used to search for planetary nebulae at
the positions of weak radio sources de-
tected in these fields with the Wester-
bork radio telescope. There appears to
be an excess of such weak radio sources,
compared to the general field, and they
have the flux density typical of planetary
nebulae at the distance of the galactic
center. It is expected that many new
planetary nebulae, most of which will be
heavily reddened, will be discovered
down to distances as small as 30' from
the galactic center.

One-Millimeter Photometry of the
Crab Xebula

A study at 1 mm has been completed
by Werner, Neugebauer, and associates.

The nebula was mapped at this wave-
length with 1' resolution in twilight ob-
servations from the prime focus of the
5-meter telescope. The peak 1-mm sur-
face brightness, 15 ±: 3 Jy into the 1'
beam, is seen in the center of the nebula
at or near the position of the pulsar; the
distribution of radiation around this peak
is roughly elliptical, with half-power
width ~2'5 X 4'5, and with the long axis
lying northwest-southeast. This result is
in agreement with the size and shape of
the nebula as determined at longer wave-
lengths. This agreement has been used
to estimate the flux expected from the
faint outer regions of the nebula. On this
basis, the total 1-mm flux from the Crab
Nebula is found to be 300 d= 80 Jy.

At wavelengths longward of 1 cm, the
total radiation from the Crab Nebula is
well described by the power law Sv oc
J,- 0.26 j^^ extrapolation of this spectrum
to 1 mm predicts a total flux of 230 Jy;
the value we have determined is in good
agreement with this and does not agree
with previous reports of millimeter-wave-
length fluxes far in excess of the extrap-
olated power law. We thus find that both
the total 1-mm flux from the Crab
Nebula and the spatial distribution of
the 1-mm radiation are consistent with
expectations from radio-wavelength ob-
servations. These results, therefore, indi-
cate that the emission from the Crab
Nebula down to wavelengths as short as
1 mm can be attributed to synchrotron
emission from a single distribution of
electrons.

PULSARS

Secular Decrease in the Visible
Intensity of the Crab Pulsar

A.- thf- rf'.<^ult of a long-term program
of monitoring the visible intensity of the
Crab pulsar, Kristian has found it to be
slowly decreasing with time. The exact
amount of this dimming should provide

a test for several competing theoretical
models of the pulsar emission mechanism
which make quantitative predictions for
such an effect. The data have not yet
been completely analyzed, but prelimi-
nary results indicate an intensity de-
crease of 0.5% per year, which is about
the 10th power of the rate at which the

HALE OBSERVATORIES 141

pulsar is slowing down. This appears to recent detection of the Vela pulsar at

favor an emission mechanism proposed V > 25 mag by a group at the Anglo-

by F. Pacini, and is consistent with a Australian Observatory.

X-RAY SOURCES

The transient x-ray source A0620-00, Lasker of Cerro Tololo Interamerican

which appeared in the summer of 1975, Observatory, made a search for pulsa-

has been followed since that time by Oke. tions of the He II A4686 emission line

The most recent observations, made in in Sk 160 (SMC X-1). For one-hour ob-

November 1976 and February 1977, show serving runs on each of three nights with

the object only slightly brighter {V = the CTIO 1.5-meter telescope, upper lim-

18.4) than before the outburst. A study its (95% confidence) of the pulsed frac-

of the spectrum indicates the presence tion of line emission were approximately

of a cool star of spectral type K5 to K7 30%. The presence of the He II A4686

and probably of luminosity class V, line and its radial velocity behavior sug-

which yields a distance for A0620-00 of gest a connection with the x-ray source,

900 pc. In addition to the cool star spec- as is the case also for Cygnus X-1. The

trum, there is a weak spectrum of the implications of this limit for the mecha-

form fv = const, which can be modeled nism of He II A4686 emission are being

by adding a 10,000°K blackbody of studied.

radius (if spherical) of 1.7 X 10'^ cm With Shectman's photon-counting

and a 4000°K blackbody of radius 13.4 linear-array detector and the Mount

X 10^ cm. This is probably a small Wilson 2.5-meter coude, Petro has repeti-

residual accretion disk. Since there is tively obtained 10-second spectra of X

now strong evidence for a 7^8 binary Persei at Ha. Previous observations (by

period, the cool star does not fill its Roche Campisi et al., Mon. Notic. Roy. Astron.

lobe, and it is not obvious why mass Soc, 176, 225, 1976; Murdin et al., Mon.

transfer occurs. One possibility is that Notic. Roy. Astron. Soc, 176, 233, 1976)

the cool star is a giant, but this is un- had shown abrupt decreases in hydrogen

likely. A second possibility is that the emission with a 10-100-second time scale.

cool star is losing mass for reasons other Petro 's 24 hours of X Per Ha monitoring

than its binary character. revealed no equivalent-width changes

Wade and Oke have completed a study greater than 5%. The data are now being

of the transient x-ray source A0535-|-26 analyzed for Ha variation on the 3U

(= HDE 245770), using multichannel 0352+30 periods of 13.9 minutes and

and SIT digital spectrograph data. The 22 hours.

star, which did not change significantly An interesting new application of the
when the x-rays turned on, is of spectral modern technologies available with the
type BO and is probably of luminosity spectrographs of the Hale reflector has
class III. The reddening Eb-v = 0.80 resulted from extensive studies of A^I
and the distance 1.8 kpc. The x-ray flux, Herculis by Greenstein. This star had
even at its peak, produces so little energy become brighter, was detected as an x-ray
compared to that of the B-star itself that source by Drs. D. R. Hearn, J. A. Rich-
one would not expect to see brightness ardson, and G. W. Clark of the ]Massa-
variations correlated with the x-ray flux ; chusetts Institute of Technology, and
no variation at the 104^ x-ray period has been extensively studied at many
was found. observatories. The continuum shows

Petro, in collaboration with Dr. Barry strong circular polarization, and there is

142 CARNEGIE INSTITUTION

also a high degree of linear polarization; have some persistent sharper compo-

both are variable with phase in an orbit nents of velocity within itself. However,

whose ^HM'iod is slightly over 3 hours. The the sharp line is very intense and moves

light-curves are veiy complex, and the with only about one-tenth the amplitude

minimum, if interpreted as an eclipsing of the broad line, 180° out of phase. If

systom. is out of phase with the radial the velocity amplitudes indicated in this

velocity and with the periodic variations diagram represented material attached

of the linear and circular polarization, to the two stars in the system, the mass

Observations by Greenstein and Rich- ratio would be 6 to 1, requiring that one

stone with the Cassegrain SIT spectro- of the objects be a neutron star or black

graph indicated that the lines had rela- hole! It is, however, not probable that

tively sharp cores and some broad wings, these are genuine stellar velocities but

The Balmer continuum is strong in emis- represent instead motions of gas in the

sion. as observed with the multichannel intense magnetic field of the degenerate

spectrophotometer by Greenstein. At star, as well as in the gravitational field

Greenstein 's suggestion. Shectman made of the two stars (of which we have not

observations over a fraction of the cycle yet seen any trace). A further conclusion

at the coude spectrograph, using his to be drawn from the relative sharpness

linear detector array at the 9 A mm~^ of the lines is that the material cannot

dispersion. These immediately revealed, be too close to the highly magnetized

for A4686 of He II. that the line had a white dwarf, whose surface field of over

sharp peak and a very broad extended 10^ gauss would broaden and shift the

wing, reaching velocities up to —1000 lines completely into invisibility. Arriv-

km s~^ When the star reappeared, ob- ing at a convincing model of this fas-

servations around a complete cycle and cinating system will be extremely difficult.

parts of two other cycles w^ere made by Although not yet known to be x-ray

Sargent and Dr. A. Boksenberg, using binaries, several other eruptive or cata-

the University College London Image dysmic variables have been studied at

Photon Counting System (IPCS) in the high galactic latitudes. Some may be

coude spectrograph. These observations simply U Geminorum binaries, but it is

are being analyzed by Greenstein and interesting that they occur at great

T. A. Boroson, a former Caltech under- heights from the galactic plane. One

graduate and now a student at the Uni- ^^^^^ ^y Arp is a faint blue object that

versity of Arizona The raw data show j^ ^^j ^^^^ magnitude on the Palo7nar

that A4471 of He I and A4686 of He II ^^ '^ J ^^^ ^^^^^^^ ^^ ^^^

both contam a rather sharp strong core * ^.u +• ^ i,u ^ ^ aat

, . ., , ^ 1 Another magnetized w^hite dwari, AN

whose intensitv may vary but whose tt iv/r-- ur-xii, ix

. ^ , *^,. , , / J Ursae Maioris, much fainter than but
position varies onlv slightly, and a very ^, . *■ ', ,. atvt tt t

{ , • r u'^- I 1 -x J. • 1 otherwise resembling AM Herculis, may

broad wing of high- velocity material , %^ , , i

,, ^ ■ u 4.-L ^ ( I. }■ be an x-ray source. It has much weaker

that moves in both stages of helium ^ , -^ ,. . . . ,

.• • -i. u 4. 4-u -L Balmer continuum emission and lower

ionization m opposite phase to the sharp . . .... „

core. Figure 5 is a reproduction of a excitation emission-line spectrum from

computer plot of the IPCS material from Greenstein s multichannel spectropho-

one cvcle. It shows isophotes of the ^^"^^^ ^"^^ ^IT results. The star PG-

A4686 line as a function of orbital phase. 2337+12, discovered by Caltech graduate

It confirms in full Shectman's discovery student R. F. Green and already reported

of the high-velocity gas and shows that on by Green, Greenstein, and Boksen-

to the contour levels here plotted (about berg, has been followed by Greenstein,

200 counts per channel above the con- who observed it at a very low luminosity,

tinuumi, the broad helium line sweeps This erratic variable, which looked

from —800 to -|-600 km s-^ and may originally like a white dwarf with weak

HALE OBSERVATORIES 143

VELOCITY km s"'

-800 -400 400 800

I n

0.4

0.3

0.2

0.1

UJ
if)
<

t 0.0

o

h-

LjJ

CD
<

0.9

0.8

0.7

0.6

AM HER
X4686

AFH E D ED C B A

Fig. 5. The He II line X4686 in the x-ray magnetic binary AM HercuHs, observed
at high resolution with the coude spectrograph of the Hale reflector at Palomar Mountain.
The (vertical) phases are fractions of the 3-hour period measured from the phase of maximum
linear polarization ; velocity is plotted horizontally. The drawn lines are isophotal contours
of the emission line (with continuum subtracted). Counts for contours A to D are 200-500;
E to J, 600-1600. Note the broad feature moving with about 300 km s"^ amplitude, arising
in gas streams with a large velocity spread, and the sharp feature moving in a nearly opposite
direction with 75 km s'^ amplitude. The x-ray eclipse occurs near phase 0.14 when the gases
producing the sharp component are farthest from the observer.

hydrogen emission lines, at minimum re- bright is concealed by an accretion disk,

sembles a low-metal-abundance sdK star When space observations are possible to

in energy distribution. At minimum it lower x-ray flux levels, many such ob-

has the color and absorption lines of a jects may well prove to have x-ray

quite cool object, which when the star is emission.

144

CARNEGIE INSTITUTION

SUPERNOVAE

The supernova search with the 46-cm
Schmidt telescope was conducted by
Kowal and Wade, under the supervision
of Sargent. In August 1976, a 14th mag-
nitude supernova was found by Wade in
NGC 5427. This was onlv the second

supernova to be found in any of the
program's Sc I galaxies in the past six
years. Forty Sc I galaxies are included
in this search.

A spectrum by Oke shows the super-
nova to be of Type I.

THE GALAXY: ITS HALO AND GLOBULAR CLUSTERS

^fetal Abundances in Globular Clusters

Searle and Zinn have completed their
program of mapping the spatial distri-
bution of globular clusters of different
metal-abundance with the Galaxy (see
Year Book 75, p. 299). They have ob-
tained accurate metal abundances for 19
clusters. The major result of this study
is that some distant clusters such as
Palomar 5, NGC 6229, and NGC 7006
are not as metal deficient as had been
thought and as would be expected if a
gradient in metal abundance existed in
the Galaxy's outer halo.

Zinn and Searle have also found that
the so-called second parameter effect (the
fact that the morphology of a cluster's
color-magnitude array depends on some
factor in addition to metal abundance)
is closely related to a cluster's location
within the Galaxy. Among clusters of a
particular metal abundance, those w^ith
bluer horizontal branches are more
tightly bound to the Galaxy than those
with red horizontal branches. The second
parameter is thought to be either an
abundance effect or an age effect. Be-
cause a gradient in the abundance of the
hea\'y elements is not found in the halo,
Zinn and Searle think it unlikely that
there is a gradient in He, C, N, or
(which are the relevant elements in this
context!, and therefore lean toward the
view that the second parameter must be
the age of the clusters.

Trying to understand these new results,
Searle and Zinn have been led to a hypo-

thetical model for the formation of the
Galaxy's system of globular clusters.
They imagine cluster formation proceed-
ing before the Galaxy itself formed in
an initially gaseous protogalaxy contain-
ing large and growing density fluctu-
ations. These fluctuations or fragments
undergo independent chemical evolution.
Cluster formation begins in massive frag-
ments near the center of the protogalaxy
where, initially, metal-poor clusters are
formed. As the protogalactic gas in these
central fragments becomes progressively
enriched in metals (as the result of
supernovae explosions), more and more
metal-rich clusters form. All the clusters
formed in these central massive frag-
ments eventually collapse to form the
tightly bound system of clusters of the
Galaxy's central bulge. The outer parts
of the protogalaxy take longer to collapse
because of the general Hubble expansion
of the LTniverse. Cluster formation even-
tually occurs in the less massive frag-
ments of the outer protogalaxy. The
supernova explosions that occur in these
disperse the gas before much enrichment
can occur, so that only relatively metal-
poor clusters form from them. Individual
variations in the enrichment history of
different fragments during this era of
halo-cluster formation, together with
their free-fall collapse, explain why an
abundance gradient is not found in the
loosely bound clusters of the Galactic
halo. To explain the phenomena associ-
ated with the second parameter, the time
difference between formation of the bulge

HALE OBSERVATORIES 14o

and halo clusters must be roughly 10® <t(^Y) < ± 0.08 if (7) = 0.3,

years. and

As part of a continuation of these in- u{hZ/Z) < ± 0.16.

vestigations, Zinn has developed a method These are very small limits. Hence, dif-

of measuring the metal abundance of ferent parts of the Ml 5 protocloud could

globular clusters from photometry of not have been enriched significantly

their integrated light. Zinn's method, differently in metal or in helium abun-

which is reddening-independent, yields a dance than other parts during the proto-

difference of roughly 0.5 mag between cloud collapse and fragmentation in the

the metal-poor cluster M92 and the formation process.

metal-rich cluster M71. So far more than A secondary result of these faint pho-

60 clusters in the Galaxy have been tometric studies is that lists of B and

measured, primarily with the 1 -meter Y magnitudes to very faint limits are

Swope Telescope at Las Campanas. available for large numbers of stars in a

compact area (22 □' for the M15 fieldj
for M15 and M92. These fields should

M15 Main-Sequence Photometry .^^^ ^^^^.^^ ^l ^^'^ ^f f ^ ^,°^ ^^""fT'^'^l^^

television-type detectors oi the Ibli,

Faint B and V photometry of many SIT, SEC, or CCD types where linearity

stars in the globular clusters M15 and and homogeneity over the field require

M92 has been completed by Sandage and calibration.

Katem in an effort to determine the in- A by-product of the M15 photometry
trinsic width of the main sequences. The is a catalog of very faint (V > 18), very
photometry covers the range 18 < F < red (\_B — F] > 1.4) field stars in the
22, 18 < B < 23 in both clusters and is 22 Q' area. These must be foreground
based on photoelectrically calibrated se- red dwarfs. Their number per square
quences and five photographic plates in degree per unit magnitude interval pro-
each color. vides data necessary to model the density

The result for M15, published during gradient perpendicular to the galactic

the report year, show a well-defined main plane for such dwarf stars that are in

sequence with a turnoff at 7 — 19.0; the the range +8 < M^ < +12. The num-

observed width in the color-magnitude ber of stars in the M15 field is so small

diagram can be entirely accounted for as to require a steep density gradient of

by the known errors of measurement, d logD {z) / dz ^=z — 1.4 (^ in kpc), similar

The errors were determined from the to that required to explain Becker's

internal scatter of individual values for photometry of very faint stars in SA57

the five plates in V and B for each star and similar to Weistrop's results in the

in each color in each half-magnitude bin interval 300 < 2 < 2000 pc. This gradi-

from F = 18 to F = 22. ent corresponds to an i folding scale

Because the main-sequence position is height of 310 pc, showing that the fore-

a function of the helium abundance (F), ground M stars are members of the old

and because the subgiant-branch position disk population rather than the extended

is a function of the metal abundance halo where the gradient is so much more

(Z), intrinsic widths of these two se- shallow at d log D{z)/dz — —0.1 {z in

quences put limits on permissible varia- kpc). A more extensive catalog of faint

tions of Y and Z in M15 itself. Observed red stars is being prepared as a part of

limits on the widths, together with the the faint Selected Area halo mapping

equations that describe sequence position project begun during the report year and

as functions of Y and Z, give the observa- described under ^^Quasars and Quasi-

tional limits on any chemical inhomo- Stellar Objects." Analysis of the larger

geneity in the M15 collapsing protocloud: material is expected to give better values

146

CARNEGIE INSTITUTION

for the disk gradient and for the disk-
halo popiihition ratio.

Color Magnitude Diagram of
Two Remote Halo Clusters

Color-magnitude diagrams of globular
clusters in the galactic halo give infor-
mation on metal abundances, and hence
data on details of the well-established
systematic metal-abundance difference
between the disk and halo of our Galaxy,
Studies of metallicity as a function of
distance from the plane can test for the
presence or absence of a gradient in
metallicity within the halo and/or the
spottiness of the enrichment process dur-
ing the halo collapse.

NGC 5053

With these problems in mind, Sandage,
Katem. and Dr. Harold L. Johnson of
the Steward Observatory prepared for
publication their 1963 study of the dis-
tant halo cluster XGC 5053. The color-
magnitude diagram was obtained using
a photoelectric sequence of 37 stars in
the interval 9.7 < V < 17.1 and from
photographic measurements of three
plates in each color. The horizontal
branch is at T^ = 16.63. The reddening is
small for this verv high-latitude cluster
(6 = + 79°) at EiB - V) = 0.01 ±
0.02, which, with M.ARR) = + 0.6
assumed, gives a distance of 15.8 kpc
from the sun and a height above the
plane of Z = 15.6 kpc.

The intrinsic color of the subgiant
branch at the level of the horizontal
branch is (B — F)o.,, = 0.68 ± 0.02
which is the bluest known value for any
well-observed cluster. Because this color
is strongly dependent on metal abun-
dance, the blue value shows that NGC
5053 is among the most metal-poor
clusters known.

The large linear diameter shows that
the cluster has never been closer to the
galactic center than ^15 kpc, since
otherwise the galactic tidal field would

have stripped the outer stars and reduced
the linear diameter. Therefore, the very
low metal abundance must be representa-
tive of the halo material out of which this
particular cluster was formed at this very
large perigalactium distance or beyond.
NGC 5053 is an intrinsically faint
globular cluster in the Galaxy at M.^^ =
—6.2. However, star counts show that
the luminosity function is normal (com-
pared with M3), and that the reported
giant-poor nature of the cluster is due
to the small number of stars rather than
to any abnormality.

The Remote Serpens Cluster Palomar 5

A very faint globular cluster was first
discovered at ^50 = 15^ IS'PS, 850 = 00°05'
by Baade on 1.2-meter Schmidt plates
before the National Geographic-Palomar
Sky Survey began, and is like a dozen
such new clusters discovered during the
survey that are among the intrinsically
faintest globular clusters known. The
color-magnitude diagram for Palomar 5
was obtained by Sandage and Hartwick
in 1969 as part of their halo-mapping
project, and was prepared for publication
during this report year.

The CM diagram is based on a UBV
photoelectric sequence of 29 stars in the
interval 13.4 < V < 18.6 and on two
photographic plates in B and in V colors.
The horizontal branch at V = 17.35,
with E(B - V) = 0.03 and MARR)
= -fO.6 assumed, gives a distance of
21.5 kpc from the sun and a height of Z
= 15 kpc above the plane (the same as
NGC 5053). However, the metal abun-
dance of Palomar 5 is not as low as that
of NGC 5053. The unreddened color of
the subgiant branch at the level of the
horizontal branch is (B — V)o,g = 0.82
it 0.02, which corresponds to an inter-
mediate metal abundance of only a factor
of 30 down from the sun rather than the
factor of -200 for NGC 5053, M15, M92,
etc. This surprising result is the same as
obtained for Palomar 5 by Hesser, Hart-
wick, and McClure and by Zinn and

HALE OBSERVATORIES 147

Searle using different methods, and hence stars in each of these systems were ob-

appears to be well established. tained with Lynds' 'Tiold" spectrograph

This is an indication that the metal equipped with a two-stage image tube.

abundance is not uniformly low for re- The spectrograms, of dispersion 100 A

mote halo objects, and therefore that the mm~^, were lightly exposed and 2 mm

relation between metal abundance and wide. The sky was observed at the same

height has a distribution. To judge from time as the object. Radial velocities were

the results of Zinn and Searle, the distri- obtained for all these objects by digitiz-

bution may not even be systematic. ing the spectra on the Kitt Peak PDS

The work on Palomar 5 by Sandage microphotometer. The sky was then sub-

and Hartwick suggests that the chemical tracted and the resulting spectra were

enrichment of the halo was spotty, caused cross-correlated with spectra of K-giant

rather by random enrichment events than stars in M15 and M92 whose velocities

by a systematic uniform enrichment. The had been accurately determined by Gunn

stirring of enrichment products within and Griffin.

the halo must therefore have been quite The resulting velocities for the 12 sys-

inefficient during the free-fall time of tems were then analyzed in conjunction

the collapse, so that the halo was not with existing velocities for the globular

mixed with much spatial uniformity clusters in the inner halo in order to esti-

during its formation, and a wide distribu- mate the total mass of the Galaxy out

tion (—2.3 < [Fe/H] < —1.0) of metal- to a radius of 100 kpc. A firm result is

licites resulted at any given halo height, that the dispersion in radial velocity

Palomar 5 has one of the faintest stays sensibly constant out to this large

absolute magnitudes for any known radius. This leads to a mass distribution

globular cluster in the Galaxy at M^ = for the outer parts of the Galaxy, which

—5.0, as determined from star counts is W (r) oc r where Tl (r) is the mass in-

and a fitting of the resulting luminosity side radius r. An accurate estimate of the

function to that for M3. actual mass of the Galaxy depends on a

knowledge of the shape of the velocity

The Outer Globular Clusters ^^'"^f^ ^^^^^^ ^J^^ peculiar motions

and Satellites ^J ^^^ systems^ This mass could be as

high as 7 X 10^^ or as low as 4 X 10^^

In 1975, Sargent, together with Dr. 'j^fl^ out to 60 kpc. The velocity measure-
David Hartwick of the University of ments also do not agree with Lynden-
Victoria, began a program to determine Bell's hypothesis that several of the outer
the radial velocities of the distant globu- systems belong to the ''Magellanic
lar clusters in the halo of the Galaxy Stream" discovered by Mathewson and
and of the dwarf spheroidal satellites of his collaborators at 21 cm.
the Galaxy. These systems are very The spectrograms of the outer halo
thinly populated with stars, so that it is systems were also studied by Dr. Anne
necessary to determine velocities for in- Cowley of the University of Michigan,
dividual stars of average brightness 18-19 together with Hartwick and Sargent, in
mag rather than to use the integrated order to investigate the metal abundance
light. To this end, observations of stars of the outer halo stars. The spectrograms
in 12 systems were made at the Kitt were compared with those of K giants in
Peak 4-meter telescope in October 1975 the metal-poor clusters M15 and ]M92
and April 1976: the globular clusters and also with similar spectrograms of K
NGC 7492, Palomar 1, 2, 3, 4, 11, 12, 13, giants in the metal-rich globular cluster
and 14 (the Arp-van den Bergh cluster), M71. The surprising result was that the
and the dwarf spheroidals Draco, Sculp- metal line strength is not correlated with
tor, and Ursa Minor. Spectra of two distance from the galactic center for these

1-iS

CARNEGIE INSTITUTION

outlying systems. Instead, the spectra
show a range of metal deficiency with
Mlo and M92 at the metal-weak end of
the range. This result is in agreement
with work by Searle and Zinn on the
inner halo of the Galaxy; it has profound
significance for theories of the formation
of the Galaxy.

Molecular Hydrogen in Galactic Sources

The discovery of molecular hydrogen
emission in Orion by T. N. Gautier III
et al. [Astrophys. J. [Lett], 207, L129,
1976), spurred intense activity. S. Beck-
with. a graduate student at Caltech, has
carried out observations of molecular
hydrogen emission in the Orion Nebula
since October 1976, in collaboration with
Persson. Becklin. and Neugebauer. An
Ebert-Fastie spectrometer has been
adapted for use at 2 ixva with an InSb
detector. Sensitivity is sufficient to make
detailed spectral studies of H2 emission
in two sources. Orion and NGC 7027.
Maps have been made with 13" and 15"
spatial resolution of emission from H2
in the r = 1 -^ S(l) transition around
the Kleinmann-Low Nebula. The low-
resolution map shows emission extended
by almost 1" with two regions of emis-
sion maxima. From this map it has been
determined that the luminosity in the one
H2 line is 2.5 L©, with a total luminosity
of ~50 Lq in all H2 lines. The emission
is intense and the intensity places strong
constraints on current models for its
source.

The map with 5" resolution shows a
number of emission peaks as well as
extended ridges and "holes" of emission.
The high-resolution map locates the re-
gions of maximum emission (presumably
the result of density variations in the
source) and displays the sizes very ac-
curately. With more detailed modeling of
the source, the interesting possibility
exists that one may be able to determine
accurately the densities in a region
around the Kleinmann-Low Nebula and
compare these w^th the densities needed

for gravitational collapse. Measurements
of various transitions other than f =: 1
-» S(l) have been made at various
positions selected from the 5" map. These
measurements are the most accurate of
any yet published and show that if the
level populations are governed by a
Boltzmann distribution, then the H2
exists as a very hot {T - 2000°K), thin
(A'ho '^ 10^^ cm-2) layer of gas. Such
measurements can detect departures of
the level populations from thermal equi-
librium; they have already placed strong
constraints on proposed models.

Galactic Center Observations

A study of compact sources near the
center of the Galaxy was brought to
fruition in a series of four papers by
Becklin, Matthews, Neugebauer, Willner,
and Wynn-Williams. In two of these
studies, high-spatial-resolution measure-
ments in the broadband filters from 1 to
10 /tm, as well as in the [Ne II] line,
were analyzed to elucidate the nature of
the compact sources in the center; these
results have been described in previous
annual reports {Year Book 73, p. 140;
Year Book 7^, p. 329; Year Book 75,
p. 290).

In a third paper, new spectroscopic
measurements of the galactic center ob-
tained with the 0.9- and 4-meter tele-
scopes of Cerro Tololo Interamerican Ob-
servatory are reported. The significant
results of this study are:

1. Ionized gas, which gives about 1/30
of the total 2.2 /xm flux near the center,
is concentrated with essentially the same
distribution as the stellar content at the
center. The Bracket y line was used to
delineate the location of the ionized re-
gion within a few parsecs of the center.
The ionized gas does not show the same
degree of dumpiness as does the 10 /^m
radiation. Also, it is apparently more
restricted in spatial extent than is the
radio-continuum emission.

2. The composition of the extended 2.2-
/xm background is confirmed as consisting

HALE OBSERVATORIES 149

of stars through the presence of the CO quatcly reproduced by a model of local

absorption beyond 2.3 fxm. star formation combined with whit^i

3. Finally, the absolute extinction at dwarf cooling theory. The subdwarf B
2.2 fim was determined by a comparison stars are found to represent less than
of radio and Brackett y data to be 3.3 0.0005 of the number density of halo
mag. This is the first measurement of horizontal-branch stars, thus accounting
the absolute value of the 2.2-fim extinc- for their observed rarity in globular
tion. Combined with a knowledge of the clusters. The mean local density of sub-
extinction curve, it puts well-defined dwarf stars of (3.4 it 1.2) X 10"®
bounds on the amount of gray extinction pc~'^ makes unlikely the possibility that
in the interstellar medium. these stars have any direct evolutionary

The fourth paper of the series discussed connection with the observed Population

the interstellar extinction curve as derived I white dwarfs or planetary nebulae.
from the compact sources at the galactic

center. In particular, observations of tj ^ tz r^u- 4.- d- c
-TT^r.^ ^^ ^ , i , i , i H and K Ooiective-Frism buvvev
IRS7, which can be shown to be a late-
type supergiant, at both 2 and 10 ycm Shectman and Preston have begun an
have related the 2-jLtm extinction to the objective-prism survey for very low-
galactic center with that in the silicate metal-abundance stars in the galactic
band near 10 fxm. The ratio of the visual halo. A 16 X 16 cm interference filter
to silicate extinction was found to be in the focal plane of the 46-cm Schmidt
8 ± 3, which is less than previous deter- telescope at Palomar limits the extent
minations (13 and 22) found for other of each spectrum to ^100 A around the
samples of the interstellar material. From H and K lines at 3933, 3968 A. Because
the colors of individual sources and the the bandpass is so narrow, very long
background, it is concluded that of the exposures can be made and the problem
30 magnitudes of visual extinction ob- of confusion at a dispersion of 250 A
served to the galactic center no more mm~^ is avoided. The survey limit is
than six originate within 3 parsecs of the between B = 13.5 and 14.0 mag. The
center. vast majority of images on each film

show H and K of typical strength for

T ' u T? /• j-TT/i,-* 7-k ^ Gr- and K-type stars and are passed

Luminosity r unction of White Dwarfs .^.i^ • ^ ■, k

and Hot Subdwarfs °^«'"- ^^°f ^ ^^ conspicuously show A-

type spectra and are presumably non-
Richard Green, a Caltech graduate zontal-branch stars. A similar number of
student, continued work on a survey of F-type spectra may contain the hotter
blue stellar objects at high galactic lati- members of the moderately metal-de-
tude. A complete sample was selected, ficient subdwarf population. A very few
consisting of 115 objects with U — 5 < spectra (^1/1000) fall into none of these
— 0.5 and covering 1434 sq deg of the categories and so far have turned out
sky down to a mean limiting magnitude to be white dwarfs or nuclei of planetary
of B = 15.7. This list includes 81 white nebulae. The object of the search is the
dwarfs, 17 hot subdwarfs, and 4 quasars, occasional star in this last group which
Absolute magnitudes were determined is in fact a giant but is so metal deficient
for the white dwarfs from Stromgren that the H and K lines of Ca II, normally
photometry, from which the luminosity the strongest lines in the spectrum, are
function was derived for hot white conspicuously weak. The discovery of
dwarfs in the solar neighborhood. The such objects would place important con-
local space density is 1.43 it 0.28 per straints on the early history of nucleo-
1000 cubic parsecs for M^ < 12.75. The synthesis during the formation of the
observed luminosity function is ade- galaxy.

150

CARNEGIE INSTITUTION

GALAXIES

Coftiposition Gradients Aa'oss
Spiral Galaxies

Following Searle's discovery of com-
position gradients across the disks of
spiral galaxies, which was reported in
Year Book 69 (p. 100). much work —
both theoretical and observational — has
been devoted to this subject by many
independent, investigators. The original
conclusions have been sustained, and at-
tention has turned to the origin of the
gradients and to the detailed physics of
the ionized regions whose spectra provide
the evidence of these gradients.

In this report year, Searle has col-
laborated with Dr. G. A. Shields of the
University of Texas at Austin in an
analysis of new and highly accurate ob-
servations of emission regions in MlOl.
Searle and Shields find that in addition
to the well-established abundance gradi-
ent there is a gradient in temperature of
the ionizing stars, the hottest stars oc-
curring far from the center of the Galaxy.
They suppose that the ionizing stars
share the abundance gradient of the gas
and show that this has important effects
on the character of the ionizing radiation
emitted by these stars. When both these
effects are taken into account, detailed
models produce an excellent fit to the
new observations, and Searle and Shields
find that the 0/H abundance ratio falls
off as r~"^/2 when r, the distance from
the center of the galaxy, lies between 5
and 25 kpc. The origin of this simple
dependence is not yet understood. Simple
theories had predicted that the abundance
would fall linearly with radius, but the
observations rule this out.

Binary Galaxies

In October 1976. Sargent shared four
nights on the Hale Telescope with Dr.
Igor Karachentsev of the Special Astro-
physical Observatory, U.S.S.R. They car-
ried out a program of spectroscopic

observations of binary galaxies from a
catalogue prepared by Karachentsev
(Comm. Sp. Astrophys. Obs. U.S.S.R.
Astron. Soc, 7, 3, 1972). Sargent and
Karachentsev obtained 100 observations
with the Cassegrain digital spectrograph
of galaxies in 48 binary systems. Many
of the pairs show signs of interactions,
and many of the spectra show strong
emission lines. A preliminary analysis of
the data, carried out by Dr. Karachent-
sev, gives a mean mass-to-light ratio
(M/L) = 8.7 d= 2.8 solar units. There
is a slight difference in (M/L) between
the spiral-spiral and elliptical-elliptical
pairs, namely, {M/L)s8 = 6.7 ± 3.5 (20
systems) and {M/L)ee = 19.5 it 8.0
(4 systems). Work is proceeding on a
more extensive analysis of the data.

Compact Galaxies

Oke and Caltech undergraduate stu-
dent Tod R. Lauer have completed a de-
tailed study of the "iron" galaxy Zw
0051 + 12, using data obtained with the
cooled SIT-Vidicon on the digital spec-
trograph. Between 3700 and 6600 A, more
than 100 emission lines have been identi-
fied and measured. In addition to the
normal lines found in gaseous nebulae,
there are a great many lines of Fe II and
probably faint [Fe II] as well. A syn-
thesis of the spectrum shows that the Fe
II lines have the same profile as the
Balmer lines but different from the for-
bidden lines of [0 III], indicating that
the Balmer lines come from the high-
density region that produces the Fe II
lines.

Some years ago, Sargent {Astrophys. J.,
160, 45, 1970) carried out an extensive
spectroscopic survey of the ''compact"
galaxies listed by Zwicky. This work
yielded many objects of great interest,
including the proto-typical "iron galaxy"
I Zw 1 and the "isolated extragalactic
H II regions" I Zw 118 and II Zw 40,

HALE OBSERVATORIES 151

which were studied later in great detail galaxies from the Zwicky lists, I Zw

by Searle and Sargent. These latter ob- 1439+53 and II Zw 1622+41.
jects are members of a small class of

dwarf, gas-rich metal-poor systems that Globular Clusters in M31
are important tor two reasons, r irst, they

provide unique information on the chemi- Sargent and Kowal, together with Drs.

cal evolution of galaxies and on the pro- Hartwick of the University of Victoria

portion of interstellar helium that is and Sidney van den Bergh of the Uni-

primordial in origin. Second, they are versity of Toronto, completed the first

probably extreme examples of bursts of part of a long-term program to find

star formation in galaxies. Searle and globular clusters out to large radii (^100

Sargent invoked the idea of such bursts kpc) in the halo of M31. The ultimate

in order to explain the properties of ab- aim is to use the virial motions of such

normally blue galaxies in general. clusters to determine the mass of M31 ;

During 1976-77, Sargent was joined however, the clusters will also be used

by Dr. Daniel Kunth, on leave of absence by Searle in his studies of the chemical

from the European Southern Observa- constitution of the M31 halo,

tory, in carrying out a second spectro- The observations for this program

scopic survey of the Zwicky galaxies, consist of wide-field direct photographs

Their primary intention was to find obtained at the prime focus of the Kitt

more objects like I Zw 118 and II Zw Peak 4-meter reflector. Twenty-nine

40, and possibly objects exhibiting even fields covering the disk and regions along

more extreme properties than these arche- the minor axis of M31 were photographed

types. Accordingly, Kunth and Sargent in 1974. During the present year, the

obtained low-dispersion spectra of 130 investigators prepared for publication a

Zwicky galaxies with the P60 Cassegrain list of 355 M31 globular clusters found

spectrograph and with the Cassegrain on these plates; 124 of these clusters are

digital spectrograph on the Hale Tele- newly discovered. The distribution of

scope (these latter observations were clusters in the inner halo of M31 is

made in poor conditions as a backup to found to fall off with radius, according

more demanding work). Only 20% of to the law <^(r) ^ r~^-^, identical to that

the galaxies turned out to have emission found in the Galaxy. It is estimated that

lines despite the fact that they were the halo of M31 contains about 200

selected from Zwicky 's lists for their globular clusters beyond a radius of 1°

blue colors. Ten of them have medium (11 kpc) from the center,
to high excitation emission spectra and

are being studied in more detail. Spectro- m u i m j. • i\/rofv

, ^ ^ . , ,. , ^ , G Lobular Clusters in M 87
photometric observations were also made

of several objects similar to II Zw 40 Oke, Searle, and Dr. Rene Racine of
that had been found in other surveys, David Dunlap Observatory, Ontario,
particularly those of the Markarian have obtained absolute spectral-energy
galaxies. In the course of the Zwicky distributions of four globular clusters in
survey, three new Seyfert galaxies were M87. These clusters have visual mag-
found. Two of these have intense emis- nitudes between 19.6 and 20.6. A com-
sion lines of Fe II, together with very parison of energy distributions shows
weak forbidden emission lines. Detailed that within the accuracy of the data they
astrophysical analyses are being made are identical. It is found that the cluster
of the spectra of these objects. In addi- energy distributions can be matched ac-
tion, such analyses are being made of curately with individual stars in galactic
digital spectrograph and multichannel globular clusters. This leads to the con-
scanner observations of two other Seyfert elusion that the four ]\I87 globular

152

CARNEGIE INSTITUTION

clusters are nuich more metal deficient
than the metal-rich chister M71. about
the same as M13, and less metal deficient
than the extremely metal-deficient cluster
M92. This conclusion must be taken with
caution, since a comparison is being made
between a star system and an individual
star. A more valid result comes by com-
pariui:: the M87 clusters with globular
clusters in M31. A detailed comparison
shows an excellent match with I\131
globular clusters of type L = 7 (van den
Bergh. Astrophi/s. J., Suppl. Ser., 19,
145. 1969 >, which corresponds to a metal
deficiency of a factor 10 with respect to
the sun. Fits at L = 1 or L = 12 are
intolerablv bad.

creases outward substantially; the light
of the galaxy has a Hubble-law shape
to at least 60 kpc radius; and the only
simple dynamical models that fit the H I
rotation law, the central velocity disper-
sion, and the extremely smooth power-
law light distribution are two-component
models with a substantial dark halo.
The resulting M/L ratios are in the
neighborhood of 100, the same value sug-
gested by the virial analysis of small
groups by Gott and Turner and Kirshner,
and of the Local Group by the ''timing"
argument.

Velocity Dispersions in Galaxies

Dynamical Studies

Sargent, with former graduate student
Paul Schechter and A. Boksenberg and
K. Shortridge of University College Lon-

Dynamical studies of elliptical galaxies don, completed work on the velocity dis-
with the SIT spectrograph have contin- persions of 13 galaxies. Observations
ued this year by Gunn in collaboration made with Boksenberg's system at the
with Thuan and with P. Schechter of the coude spectrograph of the Hale Tele-
University of Arizona. The program with scope in previous years were analyzed
Thuan is a continuation of the one begun by a Fourier method that gives both the
last year to investigate the velocity dis- radial velocity and the velocity disper-
persion in cD galaxies; the one with sion of each galaxy. The new measure-
Schechter is to investigate the dispersion ments confirm a relation between the
and rotation in flattened ellipticals. There luminosity of a galaxy and its velocity
is by now impressive evidence that rota- dispersion L ^ ct^, which was found
tion and flattening do not have the ex- earlier by Drs. Faber and Jackson of
pected simple relationship one to the the Lick Observatory.
other, and that some third integral is During an unsuccessful attempt in 1976
playing an important role in the dy- to obtain radial velocities of globular
namics of these systems. A scheme for clusters in the halo of M87, a team com-
constructing models with a simple third posed of Boksenberg, Sargent, Hartwick,
integral has been devised. and Dr. Roger Lynds of the Kitt Peak

An especially interesting elliptical gal- National Observatory made extensive ob-

axy for dynamical studies is NGC 4278, servations of the spectrum of M87 itself

which was discovered recently by Faber, out to a radius of 150". These data were

Gallagher, and Knapp to have neutral obtained with the Boksenberg detector

hydrogen, and has since been shown by at the Cassegrain focus of the KPNO 4-

Knapp. Kerr, and Williams to have a meter telescope. They are now being

neutral hydrogen disk. High-resolution analyzed by graduate student Peter

SIT spectroscopy and surface photom- Young. It appears that the velocity dis-

etr\' of this system have been obtained, persion ar in M87 falls off radically, as

and dynamical models have been con- predicted by a King model. However, o-^

structed. The results indicate that the rises sharply in the central few arcsec-

mass-to-light ratio of this sysi^em in- onds of the galaxy. This behavior implies

HALE OBSERVATORIES 15

Q

that M87 contains a tight central mass })y other Westerbork observers. An ex-
concentration of about 10^^ solar masses tensive halo, present in both the radio
— which may be a black hole. continuum and neutral hydrogen, has

also been discovered around the small

Bright Spiral Galaxies irregular galaxy NGC 1569. The halo

may be related to an extensive system ot

De Bruyn has discussed radio con- Ha filaments in which the galaxy appears

tinuum observations of nearby active to be embedded. The properties of the

spiral galaxies. The observations were halos and disks are used to set constraints

obtained with the Westerbork Synthesis on the possible models for the origin,

Radio Telescope at continuum wave- propagation, and confinement of cosmic

lengths of 6, 21, and 49 cm. The high rays in galaxies,
angular resolution at 6 cm (about 7'')

and the excellent sensitivity to faint ex- ^ 4 i r i
tended emission at 49 cm have provided

for the first time a complete picture of In collaboration with Dr. A. S. Wilson

the fine structure, large-scale structure, of the University of Sussex, de Bruyn

and spectral properties of the radio con- has completed a statistical discussion of

tinuum emission of about 10 nearby the radio continuum properties of a large

spiral galaxies. sample of Seyfert galaxies observed at

The galaxies NGC 2146 and NGC Westerbork. The most important results

3079 were found to have a very bright, of this study may be summarized as

peculiar radio structure reminiscent of follows:

that observed in NGC 4258 (van der 1. The radio emission generally comes

Kruit, Oort, and Mathewson, Astron. from a region of about 1 kpc diameter

Astrophys., 21, 169, 1972). The radio centered at the optical nucleus. The radio

spectra of these galaxies show no steep- luminosity is intermediate between that

ening at the highest frequencies, indi- of normal spiral galaxies and radio

eating that the radio structures must be galaxies.

less than about 10"^ years old. This and 2. None of the galaxies shows the

other observed features are used to argue double radio structure characteristic of

that the radio structures are related to most extragalactic radio sources and, in

nuclear activity. De Bruyn is using the particular, quasars to which Seyfert

SIT digital spectrograph at the Casse- galaxies have been likened,

grain focus of the 5-meter Hale Tele- 3. The radio luminosity of type 2 Sey-

scope to investigate optical features that ferts (those having equally wide forbid-

may be related to the anomalous radio den and permitted lines) is, on the

structures. Other galaxies that may ex- average, stronger by a factor of 5 than

hibit signs of nuclear activity in the that of type 1 Seyferts (those having

radio domain, although less conclusively, very broad permitted lines and narrow

are NGC 4631 and NGC 4736. forbidden lines) . This is a surprising

In collaboration with E. Hummel of result in view of the fact that the optical

the Kapteyn Astronomical Institute at nonthermal activity is stronger in the

Groningen, de Bruyn completed the dis- type 1 objects.

cussion of the radio continuum properties 4. No correlation exists between the
of the nearby edge-on galaxy NGC 3556. radio emission and the Ha luminosity or
The most interesting feature of this nuclear ultraviolet magnitude within each
galaxy is a radio continuum halo that type group. A weak correlation was
reaches to almost 5 kpc above the plane, found between the radio luminosity and
The halo has properties similar to those the [0 III] A5007 A emission line in-
detected around NGC 4631 and NGC 891 tensity. This result is taken to support a

154

CARNEGIE INSTITUTION

model in which the radio emission and
the forbidden Hnes both originate in an
extended region in which approximate
pressure bahmce exists between the rela-
tivistic electrons and magnetic fields on
the one hand and. on the other hand, a
thermal filamentary gas. The correlation
between the 21 -cm luminosity and 10-ju
luminosity, first discovered by van der
Kruit. seems to be mainly confined to the
type 1 Seyferts but is weaker than previ-
ously believed.

To pursue the study of Seyfert gal-
axies, de Bruyn and Sargent are using
the SIT digital spectrograph at the
Cassegrain focus of the 5-meter Hale
Telescope to obtain information on the
spatial extent, ionization state, and
kinematics of the gas surrounding the
active Seyfert nucleus. A long, narrow^ slit
is used and several positions offset from
the nucleus are observed. Preliminary
results for XGC 3516, obtained in good
seeing, indicate that the region emitting
strongly in the [0 III] A4959, 5007 A
lines measures nearly 10" in diameter,
which corresponds to about 2.5 kpc
linear size.

Dc Bruyn and Sargent have almost
completed the reduction of a large set of
multichannel scans of Seyfert galaxies
taken in 1974 and 1975 with the 5-meter
Hale Telescope. In total nearly 70 gal-
axies have been observed at a resolution
of 20 A in the blue and 40 A in the red.
The data are ideally suited for a study
of the optical continua and the relation
between emission-line strength and opti-
cal colors. They have also begun a pro-
gram to monitor the spectra of a subset
of the.se 70 galaxies to collect informa-
tion on the continuum and emission-line
variability observed by Osterbrock and
others in several of these galaxies.

De Bruyn is also using the S20 pho-
tometer attachefl to the 1.5-meter tele-
scope at Palomar to obtain UBV mag-
nitudes of the nuclei anrl the extranuclear
regions of a well-defined sample of
Seyfert galaxies from Alarkarian's lists
of objects with ultraviolet excess. Many

of these objects are too faint for accurate
centering in the smallest apertures; they
will be observed with the SIT area
photometer.

Grouping Galaxies

Sulentic and Arp have collaborated
with Dr. G. A. di Tullio of Padova Ob-
servatory in a survey to identify inter-
acting double and multiple galaxies
within 1° of 99 bright (10.0 < m^g <
12.0) spiral galaxies. A total of 96 inter-
acting systems were found in the galaxy
fields compared to 49 in control fields.
The interacting systems were found to
be concentrated along the minor axis of
the average edge-on spiral, confirming an
original result of Holmberg. The second
phase of this project, now in progress,
involves spectroscopy and photometry of
the entire sample of interacting systems.

Extragalactic Observations at
1 to 10 Microns

The nucleus of the spiral galaxy IC
342 has been mapped by Becklin,
Matthews, and Neugebauer at 1.65, 2.2,
and 10 /xm with an angular resolution of
3" (80 pc) on the 5-meter Hale Tele-
scope. At 1.65 /xm and 2.2 ^m an extended
distribution of radiation is seen which is
similar in color surface brightness to that
seen in the Galaxy and nearby spiral
galaxies such as M31. This radiation is
known in the Galaxy and in M31 to
originate from late-type giant stars. At
10 /xm the nucleus of IC 342 is also
extended and is quite bright. Although
the source of the 10-/xm radiation is
unknown, it is definitely not centered at
the position of maximum stellar density.
It most likely is associated with giant
regions of ionized gas similar to the giant
H II regions such as Sagittarius B2 in
the nucleus of the Galaxy. Thus it ap-
pears that the 10-/xm radiation originates
from dust heated by young, bright early-
type stars. The spatial distributions of
these early-type stars do not follow the

HALE OBSERVATORIES

155

distributions of the late-type stars which
define the dynamical center of IC 342.
A similar situation is seen in the galactic
nucleus. The total far-infrared flux from
IC 342 has been measured on the C-141
airborne telescope with a V beam (1.5
kpc). The total luminosity observed is
about 3 X 10^ Lq, again very similar
to that observed in the same region
within the galactic nucleus. Thus the
nucleus of the galaxy IC 342 is similar
to the nucleus of the Galaxy in the distri-
bution of late-type stars, the distribution
of the regions of star formation, and the
total luminosity observed at wavelengths
longer than 50 /xm.

Persson, Dr. J. Frogel of Cerro Tololo
Interamerican Observatory, and Dr. M.
Aaronson of Harvard College Observa-
tory have continued several observa-
tional programs designed to study stellar
populations in early-type galaxies. Pho-
tometric measurements of the CO and
H2O absorption features near 2 /x in a
number of galaxies and calibrating stars
show that the light at this wavelength
is dominated by giant stars of spectral
type near M5. This result is based on
the fact that all the galaxies observed so
far display strong H2O absorption. Giant
stars, whose presence is indicated by the
CO data, must be of late type to show
strong H2O. A proper synthesis of galaxy
populations would now seem to depend
on a more complete calibration of the
indices by using late-type stars of higher-
than-solar metal abundance.

The V — K colors for a large number
of field and cluster galaxies were ob-
tained and analyzed. A basic result of
this is that the U — V and V — K colors
correlate well for E and SO galaxies. This
arises from the dependence of both colors
on several effects of metal-abundance
variations. In comparing galaxies of dif-
ferent absolute magnitude, the known
color dependence (of U — V, for ex-
ample) can thus be doubled by using U
— K. Photometric relative distance
moduli of galaxy clusters can therefore
be obtained with somewhat less sensi-

tivity to cosmic scatter and observational
errors. Tidally stripped objects, which
are too red for their absolute magnitude,
show up easily in a plot oiU — K versus
magnitude.

NGC 4151

A five-year study of the well-known
Seyfert galaxy, NGC 4151, has been
completed by Arp. This spiral is a radio
source and an x-ray source, and it has
a pointlike nucleus that varies in bright-
ness with time. Photographs of 1972
showed possible connections of NGC
4151 to companion galaxies nearby in the
field. More recently, "deep" photographs
have been acquired with the Palomar
1.2-meter Schmidt and 5-meter reflector
and the Kitt Peak 4-meter reflector.

Intercomparison and analysis of these
plates gives evidence, in Arp's opinion,
for connections and interactions between
the central galaxy and several nearby
companions. Comparison with recent
neutral-hydrogen radio maps made at
Westerbork suggests an effect of the
nearby companion galaxy NGC 4156 on
the hydrogen distribution in NGC 4151.
The companion, NGC 4156, appears to
be a peculiar spiral and has a plume of
material associated with it. Spectroscopic
study of objects in the vicinity of NGC
4151 shows a number of redshifts near
cz = 6400 to 6700 km s-^.

NGC 1199

In 1966, Arp obtained with the 5-
meter reflector a photograph of the group
of galaxies centered on the bright E
galaxy NGC 1199. An almost starlike
object about 1' from NGC 1199 was
discovered. Spectroscopic observation
showed the compact object to have a
blue emission-rich spectrum with a red-
shift of cz — 13,300 km s-^. The E
galaxy has an absorption redshift of cz
= 2600 km s~^. The original photograph,
however, suggested that the light of the
E galaxy might be absorbed by a ring

156

CARNEGIE INSTITUTION

of material around the eompaet object.
If tliis were true, it would be proof that
tile compact object is in front of the E
galaxy; this analysis again suggested to
Arp the existence of an anomalous (not
distance-related) redshift.

In the nexi 11 years, additional plates
were accumulated, including a series of
six deep Illa-J exposures with the CTIO
4-meter telescope in Chile in 1977. From
superposition printing of this series of
plates. Arp concluded that the dark ring
surrounding the compact object was real,
and that the ring was obscuring the light
of the E galaxy behind it.

In the past year at Palomar the new^
SIT spectrograph constructed by Gunn,
S. Knapp, and Oke was used to observe
the supposed dark ring around the com-
pact object. The spectrophotometric
measures suggested that the light of the
E galaxy was both reddened and ab-
sorbed by this dark ring. Associated with
this apparent dust reddening in the com-
pact object appears to be an interstellar

/v-line absorption line at the redshift
expected for a cz = 13,300 km s~^ red-
shift. Arp's interpretation is that the
high-redshift compact object lies between
the observer and the bright E galaxy
NGC 1199.

Multiplicity of Nuclei of Ellipticals

Caltech graduate student J. Hoessel,
Gunn, and J. Knapp are undertaking an
observational investigation into dynami-
cal friction processes in the brightest
ellipticals in clusters by taking short-
exposure direct red plates with the Palo-
mar 1.5-meter telescope. The clusters
being photographed are the Gunn-Thuan-
Hoessel sample of 107 nearby rich Abell
clusters for which redshifts and pho-
tometry are available. They are studying
the light distribution in the center of the
brightest ellipticals and, in particular,
are looking for multiple nuclei as evi-
dence of recent capture. So far, photo-
graphs of 42 clusters have been obtained.

CLUSTERS OF GALAXIES

''Missing Mass^' in Clusters the velocities. This technique is capable

of an accuracy of 5-15 km s~^ for the
For the past 40 years, one of the out- velocity of a galaxy that contains hydro-
standing questions in astrophysics has gen. Observations were made during the
been that of the ''missing mass" in groups report year of 294 galaxies in 18 Turner-
and clusters of galaxies. The problem Gott groups; of these, 171 galaxies were
was clearly stated by Sinclair Smith detected at 21 cm. The measurements
(Astrophys. J., 83, 23, 1936) in connec- were made with the radio telescopes at
tion with the Virgo cluster. As part of a Arecibo, Bonn, and the National Radio
large-scale attack on this problem, Sar- Astronomy Observatory,
gent collaborated with Dr. Gillian Knapp Analysis of the data is just beginning,
of the Owens Valley Radio Observatory However, it appears that most of 'the
in an extensive study of the radial veloci- groups are well-defined dynamical sys-
ties of the galaxies in the loose groups tems; for these groups the velocity dis-
defined by Turner and Gott (Astrophys. persions are about 100 km s"^. Some
J.,. Sr/7>p^ .Ser., .3^,409, 1976). The velocity groups are chance superpositions of gal-
dispersions of these groups are expected axies at radically different distances.
to be small (< 100 km s~^) and, ac- Some groups are hydrogen-rich (that is,
cordingly, Knapp and Sargent used the they contain a very high proportion of
21-cm neutral hydrogen line to determine detectable galaxies) relative to others.

HALE OBSERVATORIES

157

Velocity Field Near Rich Clusters
of Galaxies

Shectman has calculated the expected
distribution of galaxy radial velocities
near rich clusters under the Gunn-Gott
infall hypothesis. The perturbing mass
concentrated in the cluster slows the sur-
rounding Hubble expansion. Objects on
the near side of the cluster are systemat-
ically accelerated away from the ob-
server, while objects on the far side are
systematically accelerated toward the
observer. The result is an excess of gal-
axies near the cluster velocity, even at
angular distances from the cluster center
where true cluster membership is unlikely.

In the models, as the medium in which
the cluster and its surroundings evolve
is made more and more tenuous, the per-
turbation required to form the cluster
grows, and the excess of nonmember
galaxies near the cluster velocity becomes
more and more conspicuous. This effect
is unusual in that an observable param-
eter becomes more and more sensitive to
the density of the universe as the density
decreases.

Published data on the distribution of
velocities near the Coma cluster of gal-
axies indicate a low value of the cosmo-
logical density parameter, Q ^ 0.1.
Shectman is pursuing a program to meas-
ure this effect around three other rich
clusters: Abell 1904, 2065, and 2607.
More than 200 radial velocities have
already been obtained with the Palomar
5-meter telescope and the Gunn-Oke SIT
spectrograph. A preliminary analysis
again indicates low values of the density
parameter.

Velocity Anoinalies in the Virgo Cluster

A reexamination by Sulentic of all
available redshift data in the Virgo I
cluster has revealed several anomalies.
A surprising difference was found in the
radial distribution of E-SO and S velocity
dispersions. The elliptical galaxies possess
a well-defined distribution of velocities

about the mean cluster redshift at all
radii while the spiral velocities appear
to be distributfjd bimodally within 2°
of the cluster center, and similarly to the
ellipticals farther out. This suggests the
possible existence of a radial component
to the motions of the spirals with respect
to the ellipticals that might also explain
the larger velocity dispersion of the spiral
sample. It was also found that the mean
redshift of the radio-emitting spirals is
higher than for the complete spiral
sample.

Search for Evolutionary Changes

Wilkinson and Oke have completed a
study of the absolute spectral-energy
distributions of 54 brightest galaxies in
faint clusters to search for evidence of
evolutionary changes. About 15 objects
have redshifts between 2; := 0.10 and 0.21
and constitute the sample with short
look-back time. The remaining 39 clusters
have redshifts between 0.21 and 0.47.
Over the range in time corresponding to
^ = 0.10 to 0.46, B — V is not found
to change by more than 0.03 mag. Other
effects of size comparable to or greater
than any color evolution are present, and
a comparative analysis of the individual
spectral-energy distributions suggests
that one cause of this may be a disper-
sion in metallicity by a factor 4 among
the galaxies in the sample.

Search for Intergalactic Hydrogen

A systematic search for the 21-cm
hydrogen-emission line from intergalactic
hydrogen clouds in nearby groups of
galaxies has been conducted by Sargent
and Dr. K. Y. Lo of the University of
California at Berkeley. The automated
40-meter telescope of the Owens Valley
Radio Observatory and the 26-meter
telescope of the Berkeley Radio Astron-
omy Laboratory w^ere used in the search.
Extensive observations of large areas of
the M81 group, the Canes Venatici I
group, and the NGC 1023 group, cover-

Cluster Models

15S CARNEGIE INSTITUTION

ing several luiiuired square degrees of Extended low-surface-brightness images,

sky, have been made. not discernible on the Sky Survey plates,

The upper liniit^? for the parameters are found quite frequently; they could

of the intergalactic liydrogen clouds, be nearby objects,
achieved at the Owens Valley 40-meter
t-elescope. are: the hvdrogen mass
aii^ (HI) < 5 X 10^' Wo Mpc-2 if the
clouds are < 25 kpc in size; the column

density A'(H) < 1 X 10^'^ cm~- if the Dressier has completed an analysis of

clouds are > 25 kpc — if we assume 40 the recently available Leir and van den

km s~^ to be the expected line width. Bergh catalog of 1889 rich clusters of

These results should be compared to the galaxies. By comparing the distribution

parameters of the int<^rgalactic hydrogen of Bautz-Morgan types in this sample

clouds report^^ni by Mathewson and his with predictions of Monte Carlo com-

co-workers {Astrophys. J. [Lett.'\, 195, puter simulations, he was able to show

L97, 1975). They detected near NGC 55 that the distribution of the luminosities

and NGC 300 hydrogen clouds with mass of the brightest cluster galaxies is incon-

ranging from 3 X 10" Tl q Mpc"^ to sistent with a model of statistical selec-

10^** iVo Mpc~2, and iV(H) > 2 X 10^^ tion from a universal function. Specifi-

cm~-, with observed line widths of 35-40 cally, any luminosity function steep

km s~^. enough at the bright end to reproduce

Observations of selected areas in M81, the observed small scatter in the magni-

CVn I, and the NGC 1023 groups were tude of the first-ranked cluster members

also made with the Effelsberg 100-meter will not reproduce the observed distri-

telescope of the Max Planck Institute bution of Bautz-Morgan types in clusters,

for Radio Astronomy in Germany. A few and vice versa.

hydrogen clouds of mass ranging from Dressier completed a substantial part

5 X 10^ to 3 X 10^ Wq Mpc~2 were of the observations for a study of rich

detected. Some of them correspond to clusters of galaxies containing cD gal-

uncatalogued small but extended faint axies. Plates of 24 clusters taken with

unagosonthe Palomar Sky Survey plsites. the 1.2-meter Schmidt are now being

Further observations will be made with analyzed from the standpoint of the
the more sensitive maser receiver, re- spatial distributions of the cluster mem-
assembled by K. Y. Lo, using the Owens bers. Effects of equipartition and/or
Valley 40-meter telescope. However, our dynamical friction in the cD versus non-
results and the evidence in the literature cD clusters, as well as possible correla-
indicate that while neutral hydrogen tions between average cluster density
companions of mass W ?J?o or more are and cluster type, are being investigated
often found near individual galaxies, or with the aim of discriminating among
pairs of triplets of galaxies, truly isolated cD galaxy formation mechanisms. The
hydrogen clouds ( > 300 kpc from any dynamical friction mechanism of build-
galaxy) of mass > 10^ '^q are scarce, ing a cD would result, for example, in a

An optical survey of the same groups depletion of bright galaxies in the core
studied at 21 cm was begun by Sargent of the cluster, something not to be ex-
and Lo. Broad-band IllaJ plates, cov- pected if the cD is merely a result of
ering the whole extent of the groups, are special creation. The plate material will
beingobtaincfl with the 1.2-meter Schmidt eventually be scanned with the JPL-
telescope by Kowal. The plates obtained PDS system to obtain luminosity func-
are typically 2 mag deeper than the tions for these clusters.
Palomar Sky Survey plates. A large frac- Some preliminary observations of rela-
tion of the M81 group has been observed, tively nearby (z < 0.06) clusters of

HALE OBSERVATORIES

159

galaxies have been completed with the
Palomar 1.5-meter reflector. Approxi-
mately 15 clusters have been photo-
graphed, using hypersensitized 103a-O
emulsions. These limiting exposures of
1-1.5 hours with a plate scale of 15.5"
mm~^ are now being used to determine
the relative populations of different
galaxy types in the clusters and their
spatial distributions. The aim is to dis-

criminate between models of SO forma-
tion that rely exclusively on external
stripping of spiral galaxies and models
containing a combination of factors, in-
cluding external stripping, self-stripping,
and initial conditions. Higher quality
plate material for this project is antici-
pated with the forthcoming availability
of the du Pont 2.5-meter reflector at Las
Campanas.

RADIO SOURCES

SCR Identification Program

Kristian, Sandage, and Katem have
published the second in a series of papers
intended to contribute to the eventual
optical identification of all radio sources
in the SCR catalog of bright sources.
Data obtained with the Palomar 5- and
1.2-meter telescopes provide candidates
or likely identifications for 14 sources
(3C34, 169.1, 177A, 266, 274.1, 303.1,
325, 327.1, 390, 410, 435.1A,B,C, and
437). Previous identifications by others
are confirmed or new data on redshifts,
SIT photometry, or optical positions are
given for 24 sources (3C6.1, 13, 20, 43,
61.1, 68.1, 84, 123, 169.1, 173.1, 256, 265,
268.2, 268.4, 274.1, 295, 323.1, 330, 411,
427.1, 434, 437, 438, and 460).

Accurate (< 1") optical positions
were measured for a number of field stars
near each of 40 sources, and detailed
descriptions were given of each field to
facilitate further optical work. Improved
optical or radio positions have been used
to remove some previous discrepancies.
Most of the new faint identifications are
galaxies, a finding which bolsters earlier
suggestions that most optically faint
radio sources are galaxies. The only new

quasar candidates are 3C34, 3C325, and
3C435.1C.

Identifications of Radio Sources

De Bruyn, working in collaboration
with Drs. P. Katgert and A. G. Willis
of Leiden Observatory, used the 1.2-
meter telescope at Palomar to obtain a
series of deep, 14" X 14" plates of the
5C2 region (centered at 11^, +50°). This
area has recently been observed with the
Westerbork Synthesis Radio Telescope
at a frequency of 1415 MHz. More than
200 radio sources with second-of-arc
accurate positions have been catalogued
down to a limiting flux density of 6 mJy
(Katgert, Astron. Astrophys., 38, 87,
1975; 49, 221, 1976). To obtain identifi-
cations, plates were taken in the combi-
nations Illa-J + Wratten 2C, 127 —
04 + Wratten 23A, and 127 — 04 +
Red Plexiglas (no. 76-1). The plates
were calibrated with a wedge sensitom-
eter. Some of the plates are very deep and
most have good image quality. They will
be measured with the Astroscan plate-
measuring machine of Leiden Observa-
tory.

QUASARS AND QUASI-STELLAR OBJECTS

Oke and Richstone used the Multi- spatially the quasar 3C249.1 (previously
channel Spectrometer and the SIT- thought to be stellar). These observations
Vidicon digital spectrograph to resolve revealed narrow emission lines of [0 II] ,

160 CARNEGIE INSTITUTION

[0 III], and [Ne III] from a region PJwtometry and Spectroscopy in the

about 2" away from the quasar, at a 8^ and 15^ Survey Fields
redshift consistent with that of the

quasar. This is the third case of such a Radio-quiet quasars (QSOs) outnum-

phenomenon. It appears to rule out a ber radio quasars (QSS) by a very large

gravitational model for the quasar red- factor, but the statistics of the surface

shift. Simple photoionization models for density distribution of QSOs are still

the nebula — in which it is assumed that rudimentary.

"cosmic" abundance gas is exposed to In 1967, Sandage and Luyten (Astro-

the unattenuated Lyman continuum of phys. J., I4S, 767, 1967; 155, 913, 1969)

tlie quasar — fail by a large margin to began a systematic survey of QSOs in

produce the observed line strengths, six survey fields, using the 1.2-meter

Time-dependent ionization of the nebular Schmidt and the three-image technique

gas, caused either by variations of the of Haro and Luyten to attempt an im-

central quasar intensity or by shadows provement in the statistics. Catalogs of

cast on the gas by clouds near the blue objects were produced for four of

quasar, may produce a satisfactory the fields (8:48+18; 10:244-18; 11:16

model. +30; and 15:10+24), and systematic

Oke has used the multichannel spec- photoelectric and spectroscopic observa-
trometer to measure the UV flux from tions were begun to determine the nature
quasars of sufficient redshift so that the of the blue objects and the surface
Lyman jump can be measured. Of the density of the QSOs to a photoelectric
six objects so far observed, two show magnitude limit of B = 18.5.
large decreases of intensity at and below The two fields most completely studied
the Lyman jump; the remaining four to date are those at 8^ and 15^. Photo-
show no change whatsoever. The equiv- electric UBV photometry has been com-
alent width of Ly-a is approximately the pleted for about 200 objects in these
same in all these objects. These results areas. Schmidt began a spectroscopic
suggest that quasar nuclei are sur- survey for QSOs, white dwarfs, and sub-
rounded by optically thick clouds that dwarfs in 1969, based on the photometry,
are in the line of sight to the quasar only and has published the results obtained
about a third of the time. They also in 1973 in Astrophy. J., 193, 509, 1974.
suggest that two-thirds of the Lyman The original catalog positions were ac-
continuum radiation escapes from the curate to V in each coordinate, which
quasar complex, which explains the ab- although adequate for optical studies
normally low equivalent width of Ly-a was insufficient for radio searches of the
in quasars. candidates. Sandage and Katem have

A. Boksenberg and R. F. Carswell of measured optical positions of the 176

L'niversity College London and Oke used bluest (color class I) objects in the 8^

the IPCS on the digital spectrograph field with accuracies of ±1" of arc in

attached to the 5-meter Hale telescope to each coordinate. Finding charts have

measure the energy distribution of the also been prepared for both areas,

quasar 3C68.1. The observations give a Sandage has continued the spectros-

redshift z = 1.238 and indicate that the copy in both fields to obtain as complete

object is much redder than any other an unbiased sample as possible from

quasar yet studied. More recent observa- which to estimate the surface density of

tions by Neugebauer and Oke confirm the the emission-line QSOs. The 8^ field is

very red color in the visual range, but the the most complete. Combining the

color is normal in the infrared. Work is Schmidt and Sandage statistics in this

now under way to see if interstellar red- field gives a total of 39 definite QSOs

dening within the object is important. observed, out of a total spectroscopic

HALE OBSERVATORIES 161

sample of 143 objects tried. From the by spectroscopy have been found to be

statistics on variables among the 176 emission-line QSOs. The statistics on sur-

color class I objects using the independent face density to B = 18.5, although poorer

study by Dr. P. D. Usher (see section on than the 8^ field, give closely the same

Guest Investigators), at least four more values.

objects with continuous spectra are al- The new material on the positions,

most certainly QSOs. Furthermore, among charts, spectroscopy, and photometry in

the 45 color class I objects with no both fields has been prepared for publi-

spectra to date, 11 are almost certainly cation.

QSOs, based on the statistics of optical Studies have been started in collabora-
variation. Hence, in the 42.70° area, tion with Dr. A. A. Hoag of the Lowell
there are 54 objects that are definite or Observatory to carry the systematic
almost certain QSOs brighter than B = search to much fainter limits of B = 22,
18.5. Dr. Usher has determined the pho- using his converging-beam prism spec-
tometric limit relative to the photoelec- trometer at the Kitt Peak 4-meter tele-
trie values of 90 objects measured so far scope in several Selected Areas where
by Sandage in the 8^ field. standard magnitudes to this limit are
Hence, a very conservative lower limit being determined as part of a larger
to the radio-quiet QSO surface density program of mapping the galactic halo.
is 1.3 n° brighter than B = 18.5. It is

known from the spectroscopy that redder d • i,*/n o

,. , . , ; TT J TTT • XT- Bright Quasar Survey
objects of color classes II and III m the

8^ field are also QSOs. There are ^600 Caltech graduate student Richard

catalogued objects in these classes, so Green is continuing work on the Palomar

the number of QSO candidates is sub- 46-cm Schmidt survey designed to find

stantial. Furthermore, there is a pro- bright quasars over a sky area of about

nounced incompleteness effect in the edge 10,000 \Zi°- The list of ultraviolet-excess

zones of the 8^ 1.2-meter Schmidt plate, candidates generated by computer proc-

Usher's evaluation of the incompleteness essing of the survey films at the Jet

is a factor of 1.3, which gives a lower Propulsion Laboratory has led to the

limit of 70 QSOs brighter than B = 18.5 discovery of 9 new bright quasars, as

in the area, or 1.6 per square degree. well as the detection of 8 previously

Recent studies of the increase of QSO known, in an area comprising approxi-

numbers with magnitude are consistent mately 20% of the total survey coverage.

with the early estimate of ^-^6 times The final sample is expected to provide

increase per magnitude interval. If so, new insight into the form and possible

the 8^ field density suggests a growth luminosity dependence of quasar density

curve of numbers per square degree evolution, as w^ell as an excellent source

brighter than magnitude B of log N{B) of quasars for spectroscopic study from

= 0.75-13.7, which is larger by a factor space.

of 2 than the first estimate by Sandage

and Luyten in 1967. Sandage empha- Spectroscopic Observations
Sizes agam that this density is still a

lower limit because of the large bias Schmidt and Helmut Kuhr, a graduate

against detection of red QSOs by the student at Bonn University, have essen-

three-color method. tially completed spectroscopic observa-

Spectroscopy and photoelectric pho- tions of quasars with inverted radio

tometry have progressed more slowly in spectra in the S4 part of the 5000 ^IHz

the 15^ field, but a large number of survey, compiled at the National Radio

variable QSO candidates have been Astronomy Observatory and at Bonn

found by Usher, and 80% of those tested University, covering the declination range

162

CARNEGIE INSTITUTION

35 ^"^ to 70'\ Rodshift:? wore obtained for
all but two of the 29 object* that are
brighter than magnitude 20 and that
have a spectral index hirger than -|-0.2
between 6 and 1 1 cm. The average value
of T' r^,, is around 0.58 ± 0.05. This is
lower than the typical value of 0.67
found previously for quasars with steep
radio spectra, qualitatively confirming
a similar finding reported last year.
Quasars with flat and inverted radio
spectra appear to show much less evolu-
tion (a smaller density increase toward
large redshifts) than those with steep
radio spectra.

Infrared Photometry of QSS

A significant fraction of a program of
observations from 0.3 to 10 /xm of all
QSS brighter than T^ = 17 has been
completed by Becklin, Matthews, and
Neugebauer using the infrared photom-
eter at the 5-meter telescope and by Oke
using the multichannel spectrometer. Al-
though the observations are just now
being analyzed, several aspects of the
infrared-continuum energy distribution
of quasistellar objects are clear.

1. Even within the limited sample
there is clear evidence for several types
of continua.

2. In a very general sense, the flux
density Sv rises with decreasing fre-
quency; i.e., Sv oc v~^ The energy distri-
bution cannot, however, be characterized
by simple power laws even within the
range 1 to 10 /xm.

3. Although some quasars emit most of
their energy in the infrared at wave-
lengths equal to or greater than 10/xm, not
all do. In particular, a significant frac-
tion emit their maximum energy around
3 fxm (see Fig. 6). The data are being
studied with a view to understanding the
nature of the infrared radiation and, in
particular, whether there is any evidence
for thermal radiation (e.g., arising from
heated dust) as well as nonthermal radi-
ation.

2.0 -

3C66A

0.5

BL LAC (X 1/2)

3.5 2.2 1.6 1.25
X {^)

Fig. 6. Infrared energy distributions of quasars,

Absorption-Line Spectra

Since 1973, Sargent has collaborated
with Dr. A. Boksenberg of University
College London on extensive studies of
the absorption spectra of QSOs, using
Boksenberg's Image Photon Counting
System (IPCS) on the Hale Telescope.
In November 1976, a ''people's" version
of the IPCS was installed on the Anglo-
Australian 3.8-meter telescope at Siding
Spring, NSW. Sargent joined Boksenberg
and Dr. R. F. Carswell for 11 nights,
which were devoted to commissioning the
instrument. During this period, they ob-
served several bright, high-redshift QSOs
at a resolution of 1 A. These included
PKS 2126-158 iz,m = 3.4), Q0002-422
(z,m = 2.76), Q0130-403 (^em = 3.0),
Q0329-385 (^em = 2.2), Q0453-423
(2,„, = 2.66), and Q0046— 315 (^em =
2.72). About 1500 A of spectrum in the
earth's frame were observed in each case.
These observations represent the best
high-resolution material that has been
accumulated on QSOs with redshifts

HALE OBSERVATORIES

163

above z^^ = 2.5. They will lead to im-
portant information on the behavior of
QSO absorption spectra below the Ly-
man-a emission line.

In general, it has not yet proved pos-
sible to decide where QSO absorption
lines originate — whether in material
ejected at high speeds from the QSOs or
in unrelated intervening objects such
as galactic halos. Boksenberg and Sar-
gent used the IPCS at the coude spec-
trograph of the Hale Telescope in April
1977 to demonstrate that absorption
lines are produced in the spectrum of at
least one QSO, 3C232 with z^^ — 0.53,
by material in the outer parts of a gal-
axy, NGC 3067. The sources 3C232 and
NGC 3067 form one of the QSO-galaxy
pairs whose existence was pointed out by
Burbidge, Burbidge, Solomon, and Stritt-
matter in 1971 {Astrophys. J., 170, 233).
The QSO lies 1.9' away from the galaxy,
whose redshift is 2; = 0.005. Haschick and
Burke (Astrophys. J. [Lett.], 200, L1S7,
1975) showed that the 21 -cm line is ob-
served in absorption at a velocity of
1418 ± 2 km s~^ in the radio spectrum
of 3C232, despite the fact that the line
of sight to 3C232 passes 17 kpc away
from the center of NGC 3067 (an Sa
galaxy) on the plane of the sky. Boksen-
berg and Sargent observed the optical
spectrum of 3C232 in the vicinity of the
Ca II H and K lines and at a solution
of 1 A. They found absorption lines cor-
responding to Ca II H and K at a red-

shift of 1406 ± 11 km s-^ The lines
are quite strong — K has an equivalent
width of 0.4 A. A line of this strength
would require a typical path length of
about 1.3 kpc in the Galactic plane near
the sun ; however, it has been shown that
interstellar lines of comparable strength
can be found in the spectra of globular-
cluster horizontal-branch stars in the
halo of our Galaxy.

The observations by Boksenberg and
Sargent establish that heavy elements
are present in the interstellar gas far
out in the halo of the disk of a spiral
galaxy which could produce QSO absorp-
tion lines. Furthermore, the line of sight
to 3C232 lies well beyond the Holmberg
radius of NGC 3067, so that this galaxy
presents a much larger cross section for
the production of absorption lines than
would be expected from its optical size.
Other cases of QSO-galaxy pairs will
be examined during the next observing
season.

Peculiar Quasars

Observations of the peculiar quasar
0846+51W1 were continued by Arp and
Sargent. Its brightness increased by
about 4 mag in less than one month. Sub-
sequent observations have shown low-
contrast emission lines w^hen its magni-
tude is about 20. The suspected redshift
is being confirmed and the history of its
light variations is being prepared.

OBSERVATIONAL COSMOLOGY

The Velocity Field of Nearby Galaxies

Velocities for all but 15 galaxies in the
Shapley-Ames catalog of 1249 bright
galaxies are now known and have been
collected, together with new galaxy types,
into a revised catalog by Tammann and
Sandage. The purpose is to provide a
sample to a given magnitude limit whose
biases are known, so as to permit deter-

mination of the solar motion relative to
the inner metagalaxy.

Analysis of the revised Shapley-Ames
catalog, with its nearly complete velocity
coverage, was begun by Dr. Amos Yahil
of Tel Aviv University and Tammann
in an effort to measure perturbations of
the local velocity field in the presence of
density contrasts. The aim is to assess
the role of gravity in determining the

164

CARNEGIE INSTITUTION

global world model (cf. Sandage. Tam-
inann. and Hardy. Astrophys. J., 172,
253. 1972) by iisiiii:!; the velocity per-
turbations as a measure of the local
ratio of kinetic energy to gravitational
potential energy.

But. as a first step, the solar motion
relative to the centroid of the Local
Group of galaxies must be known so as
to reduce measured velocities of all other
galaxies to a kinematic frame that has
relevance to the Universe itself, rather
than to the frame tied to the sun that
is rotating about the galactic center and
translating with the Galaxy.

Solar Motion Relative to the
Local Group

New. very accurate 21 -cm radial ve-
locities for a number of candidates for
Local Group membership were collected
from the literature and averaged by
Tammann. A solution for the solar mo-
tion relative to the complete list of pos-
sible members was made by Yahil. The
results are discussed in a paper prepared
for publication (Astrophys. J. 225, 1977
November 1 issue) by Yahil, Tammann,
and Sandage.

The best-fit solar motion relative to
the centroid of the Local Group is v{0)
=: 308 ±: 23 km s~^ toward the galactic
coordinates I = 105° ± 5° and b =
— 7° zt 4°. The solution is very well
defined by those galaxies that are un-
questionablv Local Group members
rM31. M33. IC 1613, NGC 6822, and
Fornax; the Large and Small Magellanic
Clouds were not used, as they are satel-
lites of our Galaxy and hence do not
define the solar motion relative to the
Local Group itself). With a preliminary
solution for the apex and motion from
these five certain members, it became
apparent that other candidates are con-
sistent with the solution and are almost
certainly members on the basis of kine-
matics alone. They are IC 10, the Pega-
sus dwarf (DDO 216 or A 2326), the
Wolf-Lundmark-Mellott dwarf (DDO

221, A 2359), Leo A (DDO 69, A 0956),
and IC 5152. These, together with the
original five, define the final solution
quoted above.

The dispersion about the adopted
ridge-line solution is very small at o- :=
±45 km s~^ for the radial velocities. If
the membership is defined by the 11
positive members listed here, this dis-
persion specifies the mean random mo-
tion within the Local Group to within a
factor of ^ V3.

A few objects sometimes mentioned as
candidate members of the Local Group,
namely, IC 342, NGC 6946, NGC 404,
and Maffei 1 and 2, are certainly not
members. They fall 300 km s~^ above
the ridge-line solution and are clearly
expanding away from the Local Group
w^th the Hubble flow.

Possible but unlikely members, again
based on kinematics alone, are DDO 187,
GR 8, Sextans A, Sextans B, and NGC
3109. All five have positive residuals
relative to the adopted solution of about
125 km s~^ and may be the nearest
galaxies that show the cosmological ex-
pansion. It is important to test this by
determining precise distances to these
highly resolved dwarfs from detailed
studies of their stellar content (Cepheids,
brightest stars, etc.). Most are accessible
from Chile with the du Pont 2.5-meter
telescope, and a special long-range search
for Cepheids in these systems, together
with Wolf-Lundmark-Mellott and IC
5152, will be started during the 1978
observing season.

Evidence for a Positive Curvature
of the Hubble Diagram for
Brightest Cluster Galaxies

Kristian, Sandage, and Westphal are
continuing their program of redshifts
and magnitudes for brightest cluster
galaxies to extend the Hubble diagram
to large redshifts. In the second paper
of a series, they present new photometry
for 33 clusters and new redshifts for 50
clusters, extending to z = 0.75. The new

HALE OBSERVATORIES

165

data, combined with earlier samples,
show evidence for a positive curvature
of the Hubble diagram with a formal
value of ^0 (uncorrected for evolution)
of 1.6.

The change of measured colors with
redshift agrees with Whitford's standard
galaxy predictions as far as 2 = 0.25
for 5 — y and z = 0.5 for Y — R. For
larger redshifts, the B and F pass bands
move to ultraviolet wavelengths in the
galaxy rest frame that are below the
atmospheric cutoff; for such wavelengths,
data for nearby galaxies must be ob-
tained from above the earth's atmosphere.
A. D. Code and G. A. Welch have
recently given results (preprint) from
the Orbiting Astrophysical Observatory
which indicate that the energy distribu-
tions of ellipticals show a spread from
galaxy to galaxy that increases toward
shorter ultraviolet wavelengths. If uni-
versally true, this would mean that K-
corrections at very large redshifts would
be expected to show a similar spread.
The Kristian-Sandage-Westphal data
are neither very numerous nor very ac-
curate at large redshifts, but both the
B — V and V — R data suggest that the
brightest clusters lie near to the Code
and Welch results for NGC 4486, which
are at one extreme of the spread in the
OAO data. The measured cluster colors,
which at first become systematically
redder with redshift, eventually turn over
(at z ^ 0.25 for B — 7 and z -- 0.5 for
V — R) and become bluer with increas-
ing z. This strengthens the suggestion
by R. Kron that extremely distant
clusters may appear to be relatively blue.

To avoid possible difficulties with the
K-correction and with possible selection
effects for the largest redshift clusters,
Kristian, Sandage, and Westphal re-
stricted the analysis of their sample to
clusters with z < 0.4. The data in the V
and R pass bands were analyzed inde-
pendently; they give the same results.
The currently available data samples
show, for the first time, a significant de-
parture of the Hubble diagram from a

straight line, with persistent indication
of a positive curvature.

If no evolutionary corrections are ap-
plied, the formal best-fit value for q^
is 1.6 ± 0.35. The dispersion in the abso-
lute magnitudes of the brightest cluster
galaxies continues to be strikingly .small:
the best-fit values are Mv = —23.28 and
Mu = —24.09, with a standard devi-
ation of 0.04 mag and a dispersion of
0.28 mag in each pass band.

The effects of the standard corrections
on the value of Qo were also investigated
for this datum sample. The color data,
as discussed above, support the use of
the standard Whitford X-corrections to
V and R magnitudes at least as far as
z ^ 0.5. Earlier discussions of the cor-
rections for cluster richness and Bautz-
Morgan contrast class are essentially un-
changed by the new data. The effect of
the richness and B-M corrections is to
decrease both the dispersion in absolute
magnitude and the computed value for
Qo'. if the corrections are not applied, the
best-fit value for Qo increases to 2.2.

The aperture correction, which reduces
measurements at different z's to the same
intrinsic size at the galaxies, involves a
dilemma in principle. To apply the cor-
rection properly, one should know the
true value of ^0 in advance in order to
compute the correction. Detailed investi-
gation showed, however, that the sensi-
tivity to Qo is sufficiently small so that
an iterative procedure may be used to
converge quickly to a self-consistent
single best value of Qq. The basic reason
for the relative insensitivity of the aper-
ture correction to Qo is that the standard
metric size chosen is so large that the
change of total magnitude with aperture
is slow; i.e., the growth curve is rather
flat.

After a lapse of nearly fifty years since
Hubble's original discovery of the red-
shift-distance relation, and its extension
by a factor of 200 in distance, the data
appear to be near to fulfilling their
original promise as a possible test for
cosmological models. The viability of

166 CARNEGIE INSTITUTION

this approach for finding (jo is now en- ciple. Plate material for the sample has

tirely dependent on the outcome of efforts been obtained by John G. Hoessel, a

to understand the nature and size of Caltech graduate student, and will be

gahixy evokition eftVcts ^luring the look- used for surface photometry of the

back time. To consider two opposite brightest ellipticals to study the effects

cases, the present data show that if the of dynamical friction and the universality

universe is nearly empty (^o ^ 0), then of the ^'growth curve" for such systems,

gah^xies nuist have been about 1 mag The evolutionary changes brought

brighter at z = 0.4. On the other hand, about by dynamical friction are of very

if there has been no net brightness evolu- great interest for observational cosmol-

tion since z =: 4, then the data show that ogy, but observational tests are difficult,

the universe is firmly closed, finite, and One such test is the observation of the

oscillating. structure and frequency of binary gal-
axies in clusters. It is predicted that such
systems should arise from frictional in-

The Hubble Diaqram teractions but should be short-lived; the

end result is a merger to form one larger
The survey by Gunn and Oke for faint system. A study has been undertaken
clusters of galaxies has continued this by Gunn, graduate student John Hoessel,
year with somewhat reduced effort, the and Dr. Gillian Knapp to obtain surface
emphasis having been on spectroscopy photometry and velocity data on all the
with the SIT spectrograph. The format binary central galaxies in the Gunn-
of the survey has been changed some- Thuan sample. The plate material is
what owing to improvement in plate ma- roughly half complete, and the spec-
terials. Nitrogen-baked and hydrogen- troscopic observations barely begun, al-
soaked 098-04 plates made with the though some material exists from the
Wynne field-corrector at the prime focus Gunn-Thuan redshift program,
of the 0-meter Hale Telescope reach It has been speculated that the giant and
about as deep in a 20-min exposure as do supergiant galaxies found as the central
image-tube plates. The direct plates are members of great clusters arise through
also of much higher astrometric quality, frictional accumulation of small galaxies
and it is expected that the survey will be in the cluster. A graphic example of this
conducted with them henceforth. Thirteen process may be evident in the central
such plates have been obtained this year galaxy in Zw 01 0257+35, in which many
under conditions of good to excellent tidally stripped nuclei of high surface
seeing, with the discovery of four distant, brightness are circulating in a common
rich clusters. The SIT spectrograph has envelope. The predicted frictional life-
been used to obtain redshifts for eight time of this system is very short under
additional clusters in the sample. reasonable dynamical conditions. Gunn
Acquisition of redshift data for the and Caltech graduate student Donald
Gunn-Thuan complete' Abell cluster sam- Schneider have undertaken a dynamical
pie (all Abell clusters with />) < 4, i? > study of the system. Preliminary results
1, and I 6 I > 30°, 102 in all) was fin- indicate that the light and mass of the
ished this spring. Photometry for a large central object are about the same as
part of the sample will be redone in the those of central galaxies in clusters, sug-
coming year. With these data, the low- gesting that such objects may well be the
redshift end of the Hubble diagram will progenitors of some or all supergiant
be defined as well as it can be in prin- systems.

HALE OBSERVATORIES

167

THEORETICAL STUDIES

The Abundance Distribution in the
Galactic Halo

Searle has developed a model for the
chemical evolution of the halo. He pro-
poses that initially the protohalo con-
sisted of a number of chemically isolated
gaseous fragments. Within each of these,
before the Galaxy's collapse, a small
number of globular clusters formed. In
this phase globular cluster formation is
supposed to be the main process of star

formation and chemical enrichment; the
metal abundance in each fragment, at a
particular time, is simply proportional
to the number of clusters that have, up
to that time, formed within it. Eventually
the clusters formed in these fragments
tumble together to form the present halo.
Searle has shown that random cluster
formation within these fragments leads
to a distribution over metal abundance
that is in good agreement with that ob-
served in the system of halo clusters.

INSTRUMENTATION

''Charge-Coupled^^ Detectors

With the aid of a President's Fund
grant, a collaborative program between
the Jet Propulsion Laboratory and the
Observatories to build a charge-coupled
detector (CCD) camera for direct imag-

13 A should become a nearly ideal de-
tector.

In connection with a proposal for the
Wide Field Camera on the Space Tele-
scope, Westphal and Kristian have con-
ducted a number of laboratory and
telescope tests on some experimental
ing and low-resolution spectroscopy has CCD area detectors produced in a joint

continued. This work is being done by F.
Landauer and R. Locke at the Jet Pro-
pulsion Laboratory and by Oke. The
original camera was rebuilt to (1) ac-
commodate CCDs of the latest type, (2)
put the critical electronics adjacent to
the CCD, and (3) use dry ice as the
refrigerant. The camera has been used
very successfully with the digital spec-

development program by the Jet Pro-
pulsion Laboratory and the Texas Instru-
ments Corporation. The targeted pho-
tometric and noise properties of the
detectors promise a substantial improve-
ment over previous devices. The tele-
scope tests produced excellent images of
extended objects; photometric tests of
stars agreed with photoelectric measure-

trograph in the range 6000-9000 A. The ments by Sandage to better than 2%

rms readout noise was about 250 elec- over a range of 11 mag.

trons, which is still slightly higher than

the noise introduced by the background

sky brightness when a resolution of 13

A per pixel is used. Notwithstanding this,

good spectra of quasars were obtained,

and the effective quantum efficiency was

Photon-Counting Spectrometers

A gift from Mr. Francis L. Moseley
permitted the establishment of an elec-
tronics laboratory at the Santa Barbara

over 30% between 8000 and 9000 A. If Street offices, in which Shectman is con-

the readout noise can be decreased by tinning the development of photon-

about a factor 3, which should be achiev- counting image-intensifier spectrometers,

able, the system noise will become negli- A system for the Cassegrain spectrograph

gible compared with the incoming light of the Las Campanas 2.5-meter telescope

signal, and the CCD at a pixel width of is being developed under a grant from

168

CARNEGIE INSTITUTION

the National Science Foundation. A full-
time engineer. Mr. Gary Yanik, is sup-
ported by this project.

A consortium of astronomers organized
by Shectman obtained XSF support for
the development by Reticon Corporation
of a special dual array of photodiodes
particularly suited for astronomical spec-
trometers. The two rows, each of 936
diodes, permit simultaneous measure-
ment of star and sky spectra with a
large number of resolution elements. A
dual array of this type is being adapted
to the photon-counting system described
in Year Book 75 (p. 319).

Infrared Photometer

The infrared photometer for the du
Pont telescope was completed by Persson
and Matthews. The device allows arbi-
trary placement of "signal" and ''refer-
ence" beams on the plane of the sky up
to a separation of 6'. The chopping
frequency can be higher than 40 Hz, with
an acceptable duty cycle. Two detector
systems are being tested. One is a 1.2 to
5 /x InSb system that has two filter wheels
and interchangeable detectors. The other
is a liquid nitrogen-cooled GaAs system
for use from 3300 to 8900 A. These can
be operated separately or simultaneously
or with a lO-yiA system.

Under the supervision of Gunn, work
on the coldbox for a large-format prime-
focus silicon Vidicon camera tube is near-
ing completion, and work has begun on
the construction of the electronics for
the system. It is, as of this writing, un-
clear whether the system will be sup-
planted by large CCDs as soon as it is
finished, but in either case a detector of
very large dynamic range and very high
quantum efficiency will be available
shortly for direct imaging at the Hale
Telescope. Accurate photometry to the
25th magnitude and detection deeper than
26 mag should be available in reasonable
(less than 1 hour) exposures. The elec-
tronics will also be used for the blue
detector on the tandem spectrograph

under construction ^or the Cassegrain
focus, which will presumably be a large
SIT vidicon.

A new data system for the Griffin-
Gunn radial velocity spectrometer of the
5-meter telescope has been built by S.
Knapp. The machine is now under direct
control of the PDP 11/40 computer,
which does real-time reduction of radial
velocities and displays the cross-correla-
tion traces. The software development is
being done by Zimmerman.

A data system for the Caltech 2-axis
Mann measuring engine has been de-
signed by Gunn, S. Knapp, and Hoessel.
The machine will make use of a linear
512-element Reticon array to record and
digitize 512 strips across a plate at once,
resulting in considerably faster data ac-
quisition than is possible with current
machines. The control and recording
functions will be performed by a PDP
11/34 minicomputer. The device should
be fully operational soon.

A versatile SIT system for both spec-
troscopy and direct imaging at the Palo-
mar 1.5-meter telescope is under con-
struction under Gunn's general direction,
though most of the work is being done
by the Caltech graduate students led by
S. Kent and advised by S. Knapp. The
system, which is essentially a copy of
the Gunn-Oke-Schmidt system for the
Hale Telescope, can operate in either a
256 X 256-element imaging mode or a
128 X 512 spectroscopic mode. The con-
trol system is significantly more com-
plex and sophisticated than the one at
the 5-meter telescope in order to meet
computer limitations.

Palomar Aluminizing Program

In 1976, a program was initiated to
make it possible to aluminize all mirrors
of the Palomar telescopes approximately
every two years, or more often if neces-
sary. This required many improvements.
The first step was a complete rebuilding
of the vacuum system for the 5-meter
aluminizing tank. After this was com-

HALE OBSERVATORIES

169

pleted, the 5-meter mirror was aluminized
successfully. More recently, a large crane
was added to the coude spectrograph
room to permit safe removal of the 1.2-
meter mirror of the 3.7-meter camera.
Finally, handling fixtures for various
other mirrors were manufactured. For
removal of the 1.2-meter Schmidt pri-
mary mirror, a handling truck will be
especially designed and built.

5-Meter Telescope Control Room

A new control console for the 5-meter
Hale Telescope is being assembled in

the data room on the observing floor.
This console will integrate analog and
computer functions and al.so encompass
the new control computer and the TV
guider system. All the electronic wiring
between the telescope and the computer
room, and the computer room and com-
puters and the data room, has been re-
organized in such a way that the various
cable connections become available to
any user who needs them, and the elec-
tronic setups become much easier and
obvious. A new electronics servicing room
was built next to the computer room on
the mezzanine floor.

ASTROELECTRONICS LABORATORY

With about 12 specialists headed by
J. T. Fridenberg, the Laboratory de-
signed, constructed, and installed several
systems or devices.

A new control system is being engi-
neered for the 5-meter Hale Telescope.
This system is patterned after the suc-
cessful Microprocessor Automatic Con-
trol System (MACS) developed for the
2.5-meter du Pont Telescope at Las
Campanas. Designed to offer greatly im-
proved convenience and precision of
control, it will provide for raster scan-
ning, blind offsets, variable set and guide
rates, automatic increase in the RA set
and guide rates in approaching the pole,
user-selectable epoch, and 64 user-
selected input-output functions.

Final installation of the control system
of the 2.5-meter du Pont Telescope was
completed by Laboratory personnel prior
to the telescope dedication in October
1976. Microprocessors continually up-
date the telescope position and provide
other information. The refraction cor-
rections and necessary tracking-rate cor-
rections are continuously generated. An
appropriate algorithm was incorporated
for automatic dome rotation.

The Astroelectronics Laboratory
(AEL) designed and built the basic con-

trol systems and the multiplexed control
systems for the family of auxiliary in-
struments used with the du Pont Tele-
scope. Each of these auxiliary instruments
has completely self-contained electronics
and may be operated and tested sepa-
rately. When mounted on the telescope,
auxiliary instruments can be remotely
controlled from the data room. Among
the instruments for which the Laboratory
completed new electronic systems are the
auxiliary instrument mounting base, off-
set guiders, the Cassegrain plate camera,
and the Cassegrain spectrograph.

Two projects for the Mount Wilson
Observatory received attention during
the year. The first was an electronic sys-
tem for the stellar chromosphere spectro-
graph designed by Vaughan. Essentially
a four-channel spectrometer that meas-
ures H and K line emission, together
with continuum reference intensities, the
instrument operates and prints out data
under the control of a microprocessor.

With the gift of two high-resolution
single-turn encoders from Hewlett-Pack-
ard, AEL is building a system to provide
accurate digital display of telescope
coordinates of the 2.5-meter Hooker tele-
scope on Mount Wilson. The new system
will correct the coordinates for atmos-

170

CARNEGIE INSTITUTION

pheric refraction and will also provide
readout of universal time, sidereal time,
and the calculated air mass. This system
uses a new digital clock designed at AEL
in wliich a microprocessor calculates the
sidereal time continuously from the uni-
versal time and the date. The new system
is not only less expensive than others
but has the primary advantage that one
need only set the universal time with a
standard such as WAVV; the sidereal time

is automatically calculated and con-
tinuously updated. A similar clock is
being provided for the du Pont Telescope.
AEL constructed and installed an im-
proved data system for the Oke multi-
channel spectrometer on the 5-meter
telescope. The "Multichannel II" is a
self-contained 64-channel pulse-counting
system that can be operated separately
or can be used with a PDP-11 computer
to permit real-time data reduction.

GUEST INVESTIGATORS

Dr. Saul J. Adelman of Boston Uni-
versity continued the reduction of coude
plates and spectrophotometric data taken
during previous years. He has completed
an analysis of r Herculis, B4 IV {Mon.
Xotic. Roy. Astron. Soc, in press), and
worked on o Pegasi, Al V, and Aquilae,
B9.5 III as part of a program to analyze
a number of sharp-lined normal and near
normal mid-B to F stars in a consistent
manner. Line identification studies of
several Ap stars are in progress. Study
of the energy distributions shows that
the broad continuum feature near A5200
is found in at least 90% of the Ap stars,
while it is almost absent in the normal
and HgMn stars. The feature near A4200
occurs with almost the same frequency,
while that near A6300 is rather rare.

Mr. Gonzalo Alcaino of Santiago used
the 1 -meter Swope Telescope at Las
Campanas for nine nights in June and
November to obtain direct plates for his
work on the color-magnitude diagrams
of globular clusters.

Dr. L. H. Aller of the University of
California at Los Angeles and Dr. Stanley
Czyzak of Ohio State University con-
tinued their collaborative program on
planetary nebulae. They obtained photo-
electric scanner tracings of objects such
as .1320 and NGC 2440. With the image-
tube camera built by Holland Ford of
L'.C.L.A,, they obtained excellent plates
of NGC 1535, 2022, 2371-72, 2440, and

3242 at the Cassegrain focus of the 1.5-
meter reflector in February 1977. They
also obtained high-dispersion spectro-
grams of the bright ring in NGC 7009.
They wTre able to supplement earlier
plates secured on the minor axis with a
three-night exposure on the major axis
and were able to observe a number of
faint lines with excellent resolution.
These spectroscopic observations are
supplemented by plates secured with the
image-tube camera, photelectric scans
obtained with the Oke scanner (in previ-
ous years), and some data obtained at
Lick with the image-tube scanner. A
paper on the spectrum of NGC 7009
calling attention to some of the difficul-
ties in deriving elemental abundances
was presented at the Pomona meeting of
the Astronomical Society of the Pacific.

Mr. P. Bastiaansen of Leiden Observa-
tory used the 1 -meter Swope Telescope
of the Las Campanas Observatory in
February and April 1977 in a survey for
interstellar circular polarization for a de-
tailed study. He used the Leiden three-
channel polarimeter, with pass bands of
500 to 1000 A in the visible and near in-
frared. Three stars with a high probabil-
ity (at the 3cr level) of circular polariza-
tion were found: HD 93873, 112364, and
152246.

With the 5-meter Hale Telescope at
Palomar and the 2.5-meter Hooker tele-
scope at Mount Wilson, Dr. Howard E.

HALE OBSERVATORIES

171

Bond of Louisiana State University pur-
sued a program of coude spectroscopy
of objects, most of which were originally
noted on objective-prism plates obtained
at Cerro Tololo:

1. Subgiant CH stars. These are F-
and G-type stars that appear to have
returned to the vicinity of the main se-
quence in the H-R diagram, following a
violent mixing event that occurred when
they were red giants. The high-dispersion
spectrograms confirm the enhancement of
CH and the s-process elements in their
spectra and indicate high surface gravi-
ties. Adequate material for abundance
analyses is now available for HD 88446,
HD 89948, HD 182274, and HD 216219.

2. Extremely metal-deficient red giants.
Bond's objective-prism plates have pro-
vided several new stars of this type
bright enough for coude spectroscopy.
Sufficient high-dispersion material for
abundance analysis exists for HD 2796,
HD 4306, HD 6268, HD 110184, HD
184266, HD 216143, and HD 218857.
Very preliminary indications are that
HD 4306 and HD 110184 are the most
metal-deficient stars now known, having
only about 1/1000 solar heavy-element
content. HD 184266 is also of interest as
a probable F- or G-type field horizontal-
branch star.

3. Miscellaneous. During periods of
poor seeing, high-dispersion spectrograms
of various bright stars of interest were
obtained. These included the carbon-
deficient red giants HR 885, HR 1023,
HR 6467, and HR 6766; a spectroscopic
binary that Bond is studying in collabo-
ration with G. Wallerstein, HD 214686;
and three possible RS Canum Vena-
ticorum-type binaries, HR 1099, HR
1362, and 39 Ceti.

All of the spectrograms have been
traced and measured for radial velocity,
and the abundance analyses are under
way in collaboration with an NSF-sup-
ported postdoctoral fellow at Louisiana
State L^niversity, R. Earle Luck.

Dr. A. Boksenberg and Dr. R. F. Cars-
well of University College London, col-

laborated with staff astronomers in the
use of the image photon-counting spec-
trometer. R^eferences will be found in the
sections on x-ray sources and on quasars.

During the past year, Dr. E. F. Borra
of Universite Laval, Quebec, and Dr. J.
D. Landstreet of the University of West-
ern Ontario continued to use the L5-
meter telescope at Palomar Mountain to
search for and measure magnetic fields
in Ap stars and other upper-main-se-
quence stars. This work has been done
using the University of Western Ontario
Pockels cell polarimeter with narrow-
band interference filters to measure the
circular polarization induced in the wrings
of stellar Balmer lines by the longitudinal
Zeeman effect. Magnetic curves have
been obtained by this method for the
known magnetic stars 53 Camelopardalis
and o? Canum Venaticorum (Astrophys.
J., 212, 141, 1977), HD 32633, 78 Vir-
ginis, HD 133029, ^ Coronae Borealis,
and HD 215441. Fields have been de-
tected in the Ap stars CU Virginis, (f
Draconis, and 6 Aurigae (Astrophys. J.
[Lett.'], 212, L43, 1977), and also in y
Aurigae S, 3 Scorpii, 52 Herculis, HD
170397, and HD 196178. A survey of all
northern Ap stars brighter than V := 5.0
is almost complete to a level of o-h <
300 gauss.

Drs. D. G. Currie, K. M. Liewer, R.
Braunstein, and J. Johnson of the L^ni-
versity of Maryland, and S. L. Knapp
of the California Institute of Technology
have continued the program of amplitude
interferometry observations that have
been conducted on the 5-meter and 1.5-
meter telescopes on Palomar ^Mountain
and the 2.5-meter and 1.5-meter tele-
scopes at Mount Wilson. The primary
emphasis of this program has been on
stellar diameter measurements. Recent
observations have been made on « Ceti
(which has a diameter of 0.011"), a
Tauri (0.018'0, « Orionis (0.052'0, ^t
Geminorum (0.015") , as well as o- Librae
and 8^ Lyrae and T Cephei for which
the data have not yet been analyzed.
These observations were typically con-

172

CARNEGIE INSTITUTION

ducted in two or throe wavelength re-
gions (most of which had a width of
about 100 A) to explore the general
wavelength dependence of the "hmb
darkening" and the effects of molecuhir
band structure. Most of these measure-
ments have liad an internal precision
between one and four thousandths of an
arcsecond. This precision has been em-
pirically determined by repeated observa-
tions made on succeeding nights; it has
been confirmed by comparing the value
of the diameter for stable stars measured
during different years and on different
telescopes.

A spectroscopic study of Southern H
II regions was carried out by Drs, A. C.
Danks and J. ^lanfroid of Kapteyn
Sterrewacht, Holland, using the 1-meter
Swojie Telescope with the Carnegie
image-tube spectrograph. Many of the
objects chosen were from the Rodgers-
Campbell-Whiteoak catalog. It is hoped
to obtain detailed information from these
spectra on T^, N^, ionization, and abun-
dance. Many of the objects were selected
because they are known to exhibit IR
emission. From the Balmer decrement,
an estimate of extinction can be made
from which it is hoped to obtain better
information of the gas-to-dust ratio in
these objects.

Furthermore, several prominent SO
galaxies were observed, chosen from 21-
cm observations because of their large
H I content. A .search for emission lines
was made in the nuclear region for evi-
dence of H II. None was found. The SO,
NGC 5102, was of particular interest
because dust lanes have been observed
in its nucleus. However, no emission lines
were detected.

Dr. Neal J. Evans II of the University
of Texas was unable to obtain new ob-
servations because of bad weather. How-
ever, he made progress on interpretation
and publication of results obtained
earlier. The work on infrared sources in
Monoceros R2 and on new infrared
sources associated with OH masers was
prepared for publication. This represents

the culmination of work directed at es-
tablishing the nature of a class of un-
identified OH masers emitting principally
at 1612 MHz. Sources w^ere found at 10
microns for each of 6 OH masers with
accurate positions. The variability,
energy distributions, and luminosities of
these infrared sources support the con-
tention that they are evolved stars, such
as Mira variables, with extremely thick
circumstellar shells. It is from these shells
that the infrared and maser emission
arises.

Another series of papers has the aim
of presenting a clearer picture of the
cloud conditions (temperature, density,
molecular abundances) and energetics
(heating and cooling rates for the gas
and dust). Five to six clouds have been
selected for this detailed analysis, and
work on two (S255 and SI 40) has been
completed. The other clouds are lacking
primarily the 10-20 fx data. This project
is being done in cooperation with S. Beck-
with and D. Nadeau of Caltech.

Dr. Jay A. Frogel of Cerro Tololo
Interamerican Observatory has collabo-
rated with Persson on several projects
involving infrared observations:

1. Photometric studies of composite
stellar systems. JHK magnitudes and CO
and H2O indices were measured in 50
early-type galaxies and 5 globular clus-
ters. They found that (a) the brighter
galaxies are characterized by giant-
dominated stellar synthesis models; {h)
at 2.2 /A nearly 40% of the light comes
from giants at least as late as M5; (c)
the galaxian broad-band colors tend to
redden with increasing luminosity and
decreasing aperture size; (d) in bright
galaxies the relative changes oi U — F,
V — y, and J — K as functions of radius
differ from the relative changes as func-
tions of luminosity at a fixed radius; and
ie) the integrated colors of globular
clusters are sensitive to metallicity.

2. In collaboration with Dr. Judith
Cohen of Kitt Peak National Observa-
tory, the first part of the study of giants
in globular clusters was finished; detailed

HALE OBSERVATORIES

173

results will be submitted for publication
soon. Briefly, JHK magnitudes and CO
indices for more than 60 stars in M3,
M13, M92, and M 67 were observed. The
principal findings are: (a) the infrared
magnitudes yield very accurate values
for the bolometric corrections; (6) these
bolometric corrections and V — K colors
yield effective temperatures with an ac-
curacy of ±100°K independent of sur-
face gravity or metallicity; (c) the CO
index is very metal-sensitive for metal-
poor stars and has revealed a CNO
abundance anomaly in the M3 stars; (d)
a comparison with theoretical evolution-
ary tracks results in masses near 0.6 9J? ©
and ages on the order of 10-15 X 10^;
and (e) there is a serious error in the ef-
fective temperature calibration of Schle-
singer's photometry amounting to 250 °K.

3. IR data for 25 globular clusters in
M31 has now been obtained. It has been
found that V — iv is nearly twice as sen-
sitive to metallicity changes as is t/ — V ,
and also that most metal-rich clusters
tend to have the brightest K magnitudes.
Observing runs in September 1977 should
bring this program to completion.

4. At Hale and at CTIO, multiaperture
UBVRJHK data for about 170 early-
type galaxies in the field and in the Virgo
and Coma clusters have now been ob-
tained. These cover a range of 10 in
absolute magnitude. We are now analyz-
ing these data and, in particular, are
examining the dependence of various
photometric indices on M^ and trying to
decide which index will be the most use-
ful distance indicator. Also, galaxies of
different types and locations are being
examined for differences.

Dr. T. Gehrels of the University of
Arizona continued his survey of solar
system objects with the 1.2-meter Schmidt
telescope at Palomar. The plates are be-
ing used by him and by Drs. C. J. and
I. van Houten of the Leiden Observatory.
They report that Trojans of Jupiter
occur in a peculiar distribution with
many more in the preceding Lagrangian
Point (L5) than in the following point

(JA). No evidence is found for the
existence of Trojans of Neptune (detec-
tion limit 240 kilometers radius), Saturn
(25 kilometers), and Earth (1 kilo-
meter).

Drs. R.E.M. Griffin, R. F. Griffin, and
J. Mitton of the Cambridge Observa-
tories, England, used the Mount Wilson
2.5-meter telescope and its coude spec-
trograph to obtain some of the spectro-
grams needed for a detailed comparative
study of Sirius and Vega. The study,
which is being carried out in its initial
stages by Mrs. Mitton, promises to be
of substantial interest because these two
bright stars, with nominally very similar
spectral types ( Al V and AO V, respec-
tively), have very different spectra,
Sirius being a mild Am star. The new
Mount Wilson spectra, 3 mm wide at
dispersions around 1 A mm~^ and mostly
exposed on the fine-grain emulsions
Illa-J or 127-04, allow absorption lines
as weak as 1-2 mA to be detected and
should provide a broader observational
base for the study than has been avail-
able previously.

The Griffins have begun a similar ob-
servational program to obtain high-
quality spectroscopic material on /?
Geminorum, which appears from a pre-
liminary reconnaissance to be a very
normal K giant with strictly solar abun-
dances. The analysis is to be conducted
by Mrs. Griffin in collaboration with Dr.
H. Holweger of Kiel.

R. F. Griffin, in conjunction with Dr.
G. A. Radford (also from Cambridge) ,
used the 5-meter telescope to obtain a
coude spectrogram of a ninth-magnitude
star, identified in the course of a radial-
velocity survey of North Galactic Cap
stars, which is expected to prove to be
more metal-deficient than any object yet
documented.

For the second successive season, bad
weather largely frustrated the efforts of
Gunn and Griffin in their work on the
Hyades w^ith the radial-velocity spec-
trometer on the 5-meter telescope. Never-
theless, some progress was made on cer-

174

CARNEGIE INSTITUTION

tain bright Hyades spectroscopic binaries,
and periods have now been assigned to
van Biieren 62. 121. and 162.

Drs. 0. Hanson. D. Matson. and G.
Veedor of the Jot Propulsion Laboratory
have oonipleted the manufacture and
check-out of a now three-channel in-
frared photonioter, to take full advantage
of tho rocking secondary on the Mount
Wilson 1.60-moter telescope. With this
photometer, they have obtained simul-
taneous infrared (10 /i) and visual data
on the rotational variability of a few
selected asteroids. Preliminary analysis
indicates that rotational variability is
caused more by aspherical shapes than
by albedo variations. Further analysis
will permit placing limits on the thermo-
physical parameters relating to asteroidal
surfaces and their infrared emission.

Dr. Eduardo Hardy of the Universite
Laval, Quebec, has continued his studies
of the age and chemical composition of
stars in the body of the Magellanic
Clouds, concentrating on a region 56 □'
in area, about 6° southeast of the center
of the Large Magellanic Cloud, and close
to the LMC populous cluster NGC 2209.
He has constructed a deep color-magni-
tude diagram down to T^ ^ 21.5. He used
photoelectric sequences in that cluster
published by Walker for calibrating the
plates, which were obtained at the prime
focus of the 4-meter telescope at Cerro
Tololo Interamcrican Observatory.

Hardy's color-magnitude diagram fails
to .^how any .substantial number of globu-
lar clusterlike giants. Rather, the color-
magnitude array resembles that of the
intermediate-age clusters in the Galaxy
in that it contains a distinctive feature
commonly associated with them — a well-
populated clump at M, = 0.5. In par-
ticular, the red-giant branch to which
the clump belongs is similar to that of
the intermcrliate-age galactic cluster
XGC 2158 and the LMC cluster NGC
2209. It is important to mention that
NGC 2158 is the only galactic cluster
whose color-magnitude diagram is simi-

lar to the intermediate-age populous
clusters in the LMC.

Hardy postulates that a strong burst
of star formation took place in the area
studied about 2 X 10^ years ago, and
that this burst was responsible for most
of the bright LMC members observed
there. There is also evidence for the
existence of older stars that form a
second, less-populated giant branch that
is fainter and redder than the main one.

To assess in a more direct way the
chemical composition of the red giants in
the field, several K giants distributed
over the surface of the clouds were ob-
served spectroscopically. Observations
were made at CTIO with the Vidicon
system on the 4-meter telescope. The
dispersion was 100 A mm~^ and the
spectrograms covered the region 3900-
5000 A. Giants in a number of clusters
of known age and metallicity were in-
cluded in the program to generate an
interpolation scheme. The qualitative
results from the interpolation show that
the field stars are not of the very metal-
poor globular cluster variety but resem-
ble giants of the same spectral type in
47 Tucanae. Significant differences across
the LMC or even between the two clouds
are not seen.

The combined photometric and spec-
troscopic data are then consistent with
an intermediate-age stellar population
of the nearly normal chemical abundance
in the peripheral regions of the LMC. In
the central regions, the interpretation is
less unambiguous due to the more diffi-
cult photometric problem of obtaining
deep C-M diagrams. The qualitative
spectroscopic results are not in contra-
diction with an older population of the
type discussed in Year Book 75 (p. 303).

In March 1977, Saturn was observed
by Dr. A. W. Harris of the Jet Propul-
sion Laboratory. He used direct pho-
tography through the 1.5-meter telescope
at Palomar (with a two-channel photom-
eter) to make preparations for observa-
tions of two forthcoming eclipses of
lapetus by Saturn's rings. Background

HALE OBSERVATORIES

175

sky-brightness measurements were ob-
tained near the planet over the range of
positions where the eclipse will occur, and
astrometric plates were obtained for
ephemeris improvement.

Mr. Frank Holden of San Jose, Cali-
fornia, used the 1 -meter Swope Telescope
at Las Campanas for 15 nights in meas-
uring visual double stars with a bifilar
micrometer.

Dr. J. P. Huchra of the Harvard Cen-
ter for Astrophysics continued studies of
the photometric properties of galaxies,
using the Mount Wilson reflectors to
obtain UBVR photometry of galaxies.
This is part of a project in collaboration
with T. X. Thuan and G. R. Knapp
aimed at obtaining redshifts and pho-
tometry for a complete well-defined
sample of galaxies (the Turner-Gott
sample) in the Northern Galactic Cap.

Huchra, with J. Hoessel and J. Elias
of Caltech, continued work on a near-
infrared photographic survey of the
galactic plane, using the Palomar 1.2-
meter Schmidt. A special hypersensitiza-
tion technique allows uniform, sky-back-
ground-limited EV-N plates to be
obtained in 90 minutes or less, in a
bandpass free from nebular emission and
with greater penetration through galactic
obscuration. A nearby cluster of galaxies
has been found within 4° of the galactic
plane.

Dr. Robert P. Kirshner of the Univer-
sity of Michigan used the SIT spectrome-
ter on the 5-meter telescope to obtain
spectra of six galaxies in the compact
cluster of galaxies Shakbazian 1. The re-
sulting radial velocities should provide an
improved estimate of the velocity disper-
sion of the cluster (for which there is, at
present, only an upper limit). In addi-
tion, the spectra have sufficient resolution
and signal-to-noise that an estimate of
the internal velocity dispersion of the
galaxies themselves will be derived, at
least for the brightest galaxies. This
should provide an interesting clue to the
nature of the galaxies in the cluster and

the dynamical effect of the cluster on the
individual galaxies.

Dr. John B. Irwin of Kean College,
New Jersey, using the 1 -meter Swope
Telescope at Las Campanas, has com-
pleted his first-epoch direct photographic
observations on 127-04 plates of about
55 nearby stars. The long exposures in
the red, typically four hours or more,
will be duplicated in 1977-78, and the
pairs of plates will be blinked to discover
stars of very faint absolute magnitude
that are companions of these nearby
stars.

Drs. T. Johnson, D. Matson, and G.
Veeder of the Jet Propulsion Laboratory
have continued their study of the surface
compositions of selected asteroids and
satellites, using simultaneous photometry
at 0.56, 1.65, and 2.2 /xm. Results for 30
asteroids are in press (Astron. J.). The
relative reflectances at 2.2 /xm found for
asteroids in this sample range from 0.9
to 1.7 (scaled to a value of 1.0 at 0.56
/xm). This large range and the spread
seen in the data for so-called S asteroids
indicate that the classification scheme
based on 0.3 to 1.1 /xm is not adequate.
In particular, relatively bright infrared
reflectances, Rx (2.2 /xm) ^ 7, are sug-
gestive of an optically significant metal-
lic component in the surface regolith
layers of several asteroids. The authors
have concluded that space weathering
has not significantly affected the optical
properties of asteroids; they see no evi-
dence for the production of glass or
glassy agglutinates on asteroid surfaces.

Dr. Gillian Knapp of the Owens Val-
ley Radio Observatory, with Dr. R. K.
Ulrich of the University of California at
Los Angeles, has been using Shectman's
pulse-counting spectrometer at the coude
focus of the Mount Wilson 2.5-meter and
the Hale 5-meter telescopes to measure
Ha and H^ emission-line profiles of about
20 T Tauri stars with resolutions of 5 A
mm~^ and higher. Observations of eight
stars have been obtained. By comparing
the emission profiles, it is hoped to de-
termine whether the ionized matter asso-

176

CARNEGIE INSTITUTION

ciated with these stars is due to mass
infall or outflow.

Dr. Jolin Korinendy of the University
of California at Berkeley has continued
a study of the distinct components in gal-
axy-brightness distributions {Year Book
76, p. 327 1. Altogether. 240 plates for
surface photometry have been accumu-
lated for this and related work. The
photometric data are now complete ex-
cept for a few photoelectric zero-point
measurements. Rapid scanning with the
PDS microphotometer and reduction of
the plates are planned for the winter of
1977.

Kormendy also began a survey pro-
gram to study the interaction of bars and
their related components: lenses, inner
rings (e.g.. NGC 2523 in the Hubble
Atlas), outer rings (e.g., NGC 2859), and
the nature of the spiral structure. The
121 brightest SB galaxies north of —30°
declination were examined on the Sky
Survey copy plates, supplemented by
Schmidt Illa-J plates and 2.5-meter tele-
scope plates when available. The fre-
quency of occurrence and size distribu-
tions of the various components w^ere
measured, with the following main re-
sults:

1. Of all SBO-SBa galaxies, 54% have
both a bar and a lens. There is an inti-
mate connection between these com-
ponents: in 17 of 20 cases, the bar ex-
actly fills the lens in one dimension. In
contrast, no survey galaxies of type
SBab-SBc had lenses, but 76% had inner
rings. Clearly lenses are an important
component in early-type barred galaxies,
comparable in frequency to the rings
occurring in later types.

2. These and other observations sug-
gest, but do not prove, that there is a
process that makes some bars evolve
fairly rapidly into nearly axially sym-
metric objects. The result is a lens. Since
half of all early-type SB galaxies are
transition cases, the process is secular.
If there is evolution, the more likely
direction is from bar to lens, because
bars are generally believed to form by

dynamical instabilities in the violent
phase of galaxy formation. It is suggested
that processes of internal secular evolu-
tion can dramatically alter galaxy struc-
ture.

3. The distribution of apparent axial
ratios suggests that inner rings are round.
Lenses are slightly tri-axial, with a pre-
ferred equatorial axis ratio of ^0.9 ±
0.05. Outer rings can be circular, but are
usually prolate, the shortest dimension
being the one filled by the bar.

4. The sizes of rings and lenses are
well correlated with the absolute magni-
tude of the galaxy, such that the mean
surface brightness is constant for each
morphological type. Bars and lenses have
the same size distributions.

5. The disk structure was examined to
see whether bars are associated with
global spiral patterns, usually interpreted
as density waves, or with NGC 2841-type
disks, which have many short spiral fila-
ments but no global patter. Of 61 survey
plates with spiral structure, 55 have
global patterns. The rest are transition
cases; none is purely filamentary-armed.
However, even the incomplete sample in
the Hubble Atlas shows many nonbarred
galaxies with pure NGC 2841-type struc-
ture. This supports the well-known sug-
gestion that bars strongly help to drive
a density wave.

Kormendy and Dr. T. X. Thuan of
the University of Virginia completed a
study of the diffuse background light in
the Coma cluster. The measurements
were made on isodensity tracings of 13
B,G,V, and R calibrated plates taken
with the 1.2-meter Schmidt telescope.
Vignetting produced by the vanes that
hold the plateholder was eliminated.
They report that the brightness of the
background light between 4' and 14'
radius decreases from 25.7 to 27.8 G
mag arcsec"^, and that the total light in
the annulus is G = 11.22 mag (i.e., 1.7
X 10^1 Lao for Ho = 100 km s-i
Mpc~^), which is ---'45% of the light in
galaxies alone, or ^30% of the total.
This does little to alleviate the missing

HALE OBSERVATORIES

177

mass problem. Their results suggested
that both the isodensity contours and
the equivalent profile of the diffuse light
closely parallel the distribution of light
in galaxies, implying no strong mass
segregation. However, the background
appears to be bluer than the galaxies.
This is consistent with the hypothesis
that it consists of stars tidally stripped
from galaxies, which generally become
bluer at larger radii.

Dr. W. Krzeminski of the N. Coper-
nicus Astronomical Center, Warsaw, used
both 1.5-meter telescopes at the Mount
Wilson Observatory and at the Palomar
Observatory during the period May 28-
August 8, 1976. The research was done
in close cooperation with William C.
Priedhorsky, Caltech graduate student.
They carried out multicolor {UBVR)
photometry of the x-ray binary AM
Herculis. The results suggest that the
red component of the optical flux is
closely related to the source of optical
circular polarization in the system. They
conclude from the periodic modulation
of flux in the U through R bands, which
is particularly well defined when plotted
as color curves, that the primary and
secondary minima are neither eclipses by
a secondary star nor by a hot spot. They
suggest that the primary minimum in
the visual light curve is the eclipse of a
region of intense optical emission in the
magnetic field near the surface of a
degenerate dwarf by that dwarf itself.
Results of this work are in press in the
Astrophysical Journal.

Krzeminski and Priedhorsky found
{Int. Astron. Union Circ, 2974, 1976)
that the central star of the planetary
nebula Abell 63 = IJU Sagittae is an
eclipsing binary with period 1P09™6.
The eclipse, -^4.3 mag deep, lasts 70
minutes with a flat-bottomed minimum
of about 16 minutes. Absence of light
flickering, the depth (and totality) of
the eclipse, and the spectrum (studied
by J. S. Miller, Lick Observatory, and
J. L. Greenstein) make it unlikely that
the object is an old nova.

Mr. Howard H. Lanning of Mount
Wilson Observatory continued his UBVr
[)hotometric study of seasonal variations
and the variable HZ22. Poor weather
conditions hampered the work.

Lanning also obtained spectroscopic
observations of the newly discovered
eclipsing 08 III OB subdwarf BD-
3°5357 (Dworetsky, Lanning, Etzel,
Patenaude, Inform. Bull. Var. Stars, 1284,
1977; Mon. Notic. Roy. Astron. Soc, in
press). TD-1 satellite observations show
this unique binary to be a moderately
bright UV source (38,000°K). This dis-
cordant SAO K-type classification, ob-
served UBV colors, strong Ca II, H and
K, and Ha emission lines suggested the
possible binary nature. A period of ^9.2
days was determined from the photom-
etry obtained at San Diego State Uni-
versity Mt. Laguna Observatory. The
light curve is characterized by a deep
UV eclipse of 13 hours' duration with
an ingress/egress time of only 24 minutes ;
large variations outside eclipse indicate
a strong heating effect. A preliminary
value of K = 27 ±: 3 km s"^ w^as found
from 15 plates taken with the X-spectro-
graph at a dispersion of 42 A mm~^. No
lines of the subdwarf are visible, though
continuum dilution of the G star is con-
siderable. Preliminary values for the
masses and radii of the components are:

m,a = 0.6 Wq, Esd = 0.15 Rq, mGs

= 2.5 TIq, Rg» = ^Rq, with a separa-
tion of the two stars of ^27 Rq.

A systematic search for the 21-cm
hydrogen emission line from intergalactic
hydrogen clouds in nearby groups of
galaxies has been conducted by Dr. K.
Y. Lo of the University of California at
Berkeley in collaboration with Sargent.
The automated 40-meter telescope of the
Owens Valley Radio Observatory and
the 26-meter telescope of the Berkeley
Radio Astronomy Laboratory were used
in the search. Extensive observations of
large areas of the M81 group, the Canes
Venatici I group and the NGC 1023
group, covering several hundred square
degrees of sky, have been made.

17S

CARNEGIE INSTITUTION

The upper limits for the parameters
of the intergahictic hydrogen clouds,
achieved at the Owens Valley 40-meter
telescope, are: the hydrogen mass ^iV
(HI I < 5 \ 10''O.Vc) Mpc-- if the clouds
are < 25 kpc in size; the column density
.V(H) < 1 X 10^^ cm-- if the clouds
are ?^ kpc. if one assumes a 40 km s"^
line width. These results may be com-
pared to the parameters of the inter-
galactic hydrogen clouds reported by
Mathewson and his co-workers {Astro-
phys. J. [Lett.]. 195, L97, 1975). Thev
detected near XGC 55 and NGC 300
hydrogen clouds with mass ranging from
3*X 10" a^^o Mpc-- to 10^^ d)}Q Mpc--,
and A'(H) > 2 X 10^^ cm-2, with ob-
served line widths of 35-40 km s"^.

Observations of selected areas in the
M81. CVn I, and the NGC 1023 groups
were also made with the Effelsberg 100-
meter telescope of the Max Planck In-
stitute for Radio Astronomy in Germany.
A few hydrogen clouds of mass ranging
from 5 X 10^ to 3 X 10^ WiQ Mpc-^
were detected. Some of them correspond
to uncatalogued, small but extended
faint images on the Palomar Sky Survey
plates.

Further observations will be made with
the more sensitive maser receiver, re-
assembled by Lo, using the Owens Valley
40-meter telescope. However, these re-
sults and the currently available evidence
in the literature indicate that while
neutral hydrogen companions of mass
10* Wq or more are often found near
individual galaxies, or pairs or triplets of
galaxies, truly isolated hydrogen clouds
( > 300 kpc from any galaxy) of mass
> 10* ^Q are scarce.

An optical survey of the same groups
studicfj at 21 cm was begun by Sargent
anrl Lo. Broarl-band Illa-.J plates, cov-
ering the whole extent of the groups, are
being obtained with the 1.2-meter Schmidt
telescope by Kowal. The plates obtained
are typically 2 mag deeper than the
Palomar Sky Survey plates. A large frac-
tion of the M81 group has been observed.
Extended low-surface-brightness images,

not discernible on the Sky Survey plates,
that could be nearby objects are found
quite frequently. An attempt will be
made to obtain redshifts of these objects
by searching for 21-cm hydrogen-line
emission and to resolve the stars in some
of these systems with large-scale photo-
graphic plates.

The joint radio and optical surveys
should eventually help to define the en-
vironments of galaxies in the nearby
groups.

Dr. B. F. Madore of the Institute of
Astronomy, University of Cambridge,
England, used the Swope 1-meter re-
flector at Las Campanas, Chile, for pho-
toelectric photometry during two dark
runs in November and December 1976.
Photoelectric sequences were completed
for 12 southern quasar fields to be
studied at the Anglo-Australian Tele-
scope. Also, a test of the amplitude
mapping of the Cepheid instability strip
was begun; ten 6-day Cepheids in the
Large Magellanic Cloud were monitored.
These Cepheids were chosen to have
periods within a few percent of each other
but with widely differing amplitudes.
Time-average colors and magnitudes are
being sought in order to define the slope
of the lines of constant period directly,
to determine the sense of the run of am-
plitude with intrinsic color, and even-
tually to compare these results with
similar studies in the Small Magellanic
Cloud and the Galaxy.

Dr. Bruce Margon of the University of
California at Los Angeles used the Palo-
mar 1.5-meter telescope to obtain pho-
tometry of the visible companions of two
binary x-ray pulsars, Hercules X-1 and
GX 1+4; data were successfully acquired
on each of two nights. For Her X-1,
narrow-band interference filters were
used in an attempt to map the binary
phase dependence of the A4686 emission
in the companion star, HZ Herculis, and
thus understand the origin of this emis-
sion. For GX 1+4, the objective is to
search for optical variability correlated
with the 122-sec x-ray periodicity. Since

HALE OBSERVATORIES 179

both effects are expected to be binary- in the mean velocity. The new data indi-

phase dependent, further observations cate that the center-of-mass velocity has

will be required. increased by 21 km s~^ since the observa-

Dr. D. H. McNamara of the Brigham tions by Miller and Preston were made.

Young University has pursued three spec- The orbital period of the binary must

troscopic investigations during the report be very long, certainly exceeding 20 years,

year: Prof. Edward P. Ney of the University

1. NGC 2264. A radial velocity study of Minnesota reports, in the course of
of the A- and F-type stars in the star work conducted with the 1.2-meter
cluster NGC 2264 has been completed. Schmidt telescope at Palomar, that the
These objects have been considered to discovery that the "egg nebula" (TV
be members in a state of gravitational Zwicky 67) has very highly polarized
contraction to the main sequence. The nebulosities (>50% in the visual) and
brighter objects in the cluster, mainly that these nebulosities are associated
B-type stars, have radial velocities in with a cold, central infrared source moti-
the interval of +20 to +30 km s~^. The vated a search of selected fields for
A and F stars have velocities as low as polarized objects. In August 1976 pairs
km s~^ to as high as 60 km s~^. The of plates were obtained with orthogonal
radial-velocity data indicate that many, polarizations. In all, 18 separate 4° X
if not all, of these objects are simply 4° fields were photographed; most of
foreground stars, a conclusion corrobo- these plates have been studied with a
rated by narrow-band photometry and blink microscope. Eight highly polarized
star-count data in the vicinity of the objects were discovered (in addition to
cluster. the egg nebula). The most interesting of

2. Dwarf Cepheids. An attempt was these seem to be: (1) M-2-9, an emis-
made at Palomar to investigate by con- sion-line nebulosity discovered by Min-
tinuous-trail techniques the radial veloc- kowski to have an emission spectrum
ities of CY Aquarii and XX Cygni. The like rj Carinae, is highly polarized in red
spectra obtained indicate that XX Cyg and blue light and in Ha, and has an
is the longest-period metal-poor dwarf embedded cold infrared source; (2)
Cepheid known. Because of poor observ- Parsemyan21; (3) Parsemyan 22; (4)
ing conditions, the program was changed LKHa 233; (5) IRC+10420, a star dis-
to observe the variable SS Piscium. This covered in the Caltech infrared survey,
proved to be a very useful study because which has an F supergiant spectrum in
it is clear that SS Psc is not an RRc the visible and an infrared excess like rj
variable but instead is an RRs variable Carinae. The image is almost stellar but
or dwarf Cepheid. Apparently it has the is polarized in the blue by about 40%.
longest period of any star of the dwarf Ney believes that these objects are
Cepheid group. It now appears that the stars embedded in optically thick dust
variables that have been classified as shells and that the polarization is due
RRc variables are really a mix of normal to scattering of light by dust grains.
RRc variables with masses of the order Models of preplanetary systems and of
of 0.5 5D?o and dwarf Cepheids with preplanetary nebulae suggest geometries
masses of the order of 2 SDI©- of this kind.

3. AW Persei. A series of spectrograms Dr. Valdar Oinas of Indiana Univer-
of AW Per was obtained in January at sity in June of 1976 used the Cassegrain
the 2.5-meter telescope on Mount Wilson, scanner at the Mount Wilson 1.5-meter
AW Per is a classical Cepheid with a telescope to obtain energy distributions
blue companion. Early radial velocity of 8 Scuti stars as part of a program on
data obtained by Miller "and Preston abundances.

from 1960 to 1963 had shown no change Dr. Peter Pesch, Director of the

ISO CARNEGIE INSTITUTION

Warner and Swasey Observatoiy, used major axis less than 1 A and a period of

the Mount Wilson 1.5-meter telescope in less than a year. With our Apollo asteroid

March and May 1977. As part of a con- discovery, 1977 HA, the total number of

tinning program of systematic observa- known Apollo asteroids has grown to 23.

tions of a comj^lote and kinematically This now increases by a factor of 2 the

unbiased sample of late-type M dwarfs, number of Apollos known in 1973. An-

he obtained photoelectric VJ photometry other unusual asteroid, 1977 CA, has

of 40 stars from Sanduleak's catalog. orbital characteristics similar to those

Dr. Gibson Reaves of the University objects found in the Phocaea region, but

of Southern California examined several with an unusually high inclination of 31°.

hundred faint dwarf galaxies, — 12 > Dr. Susan M. Simkin of Michigan

M^B > —15. in the Virgo cluster on a State University used the 1.2-meter

series of eleven long-exposure Illa-J Schmidt for measuring colors of the faint

plates taken by Sandage with the 1.2- extended regions of double radio galaxies;

met^r Schmidt. They appear to be gen- the program is nearing completion. Satis-

erally similar to the dwarf ellipticals in factory U,G,R plate material has been

the Local Group, but a significant frac- obtained for 3C33, 3C98, 3C390.3, and

tion of them have nuclei. A rough de- 3C405. These data are being analyzed for

scription of the areal distribution of pairs of galaxies: 3C33 paired with 3C-

these faint dwarfs may be made by 405 and 3C98 with 3C390.3. The analysis

representing the (projected) density in compares objects of similar redshift and

dwarfs per square degree, /(r), at dis- nuclear excitation.

tance r from the Sandage-Tammann Results for 3C33 and 3C405 show that
center of the cluster, by an exponential: both objects have blue ''jets" (5-30 kpc
f(r) ^ exp i—r^R). For the IC 3475- in length) emerging from their nuclei at
type dwarfs, R = 1?6; for the Sculptor directions ^^20° to the radio axes. In-
types, R = 2?2; and for the "pure" spection of Cassegrain spectra taken at
dwarf ellipticals. R = 3?5. For compari- Kitt Peak National Observatory by Sim-
son, the projected density of the normal kin and at the 5-meter telescope by
early galaxies suffers an e-fold decrease Schmidt show that these ''jets" display a
at /? = 2?9, the normal spirals at i? = normal nebular emission-line spectrum.
4:4. The significance of these results is In addition, superimposed on the outer
not yet apparent. regions of the parent cD galaxies are

The systematic search program for broad, faint (^25 mag Q'O, blue {B

new planet-crossing asteroids, using the — V ^^ 0.2) spirallike features that

46-cm Schmidt telescope, was continued coincide with the radio bridges observed

by Dr. Eugene M. Shoemaker and Mrs. in both galaxies. Finally, a diffuse patch

Eleanor F. Helin of the Division of Geo- of optical emission {B ^ 26 mag □"?

logical and Planetary Sciences. Two size 3" X 5" or 5 X 8 kpc at 344 mpc)

hundred and fifteen independent program has been found in the southwest radio

fields were photographed, plus another lobe of 3C33. The peak intensity in this

75 follow-up fields of new discoveries, optical feature lies within 0.5" of the

Sky coverage this year has led to the peak in the radio emission. Since this

disoovery of eight new asteroids: 1976 particular radio lobe is one of the strong-

QA. 1976 QB, 1976 QD, 1976 SO, 1976 est and most compact known, and since

SD. 1976 I'A. 1977 CA, and 1977 HA. it has a scintillating component at 3.7

In addition, a new comet was discovered, GHz, there is a good chance that the

Comet 1977e, Helin. Of particular note diffuse optical emission is truly associ-

was the independent discovery of 1976 ated with the radio emission.

UA, the second member of a new orbital The data for 3C390.3 and 3C98 are

class (1976 AA type), which has a semi- in the preliminary stages of reduction.

HALE OBSERVATORIES

181

The 1.2-meter Schmidt survey of Sey-
fert galaxies has only begun. Eventually
it will lead to G and R intensity maps
for all of the Seyferts in Adams' (1976)
survey with z < 0.07. This will permit a
quantitative classification of the faint
disk structures in these objects. Material
has been obtained for four objects. One
of these is NGC 4151, for which the
intensity maps show (1) that the faint
outer spiral arms are discernible in the
red as well as the blue but are broader
in the red, and (2) the object's mass-to-
light ratio increases as a function of
radius. The latter work has been done in
collaboration with A. Bosma of the
Kapteyn Institute.

Dr. Trinh X. Thuan of the University
of Virginia has used the 1.2-meter Schmidt
to obtain calibrated fine-grain plates of
the poor clusterings apparently contain-
ing supergiant CD galaxies identified
from the Palomar Sky Survey by Morgan
and his collaborators (Morgan, Kayser,
and White, Astrophys. J., 199, 545, 1975;
and Albert, White, and Morgan, Astro-
phys. J., 211, 309, 1977). Surface pho-
tometry of these giant ellipticals is
planned in order to investigate whether
they are "genuine" cDs, i.e., of the same
kind as those that occur in richer clusters.
The fine-grain emulsions will also pro-
vide reliable morphological classification
for the cluster members. Redshift data
are also being obtained (in collaboration
with Gunn) to determine cluster mem-
bership. All these data assembled should
give clues to dynamical processes (such
as dynamical friction) operating in
clusters of galaxies.

The Mount Wilson 1.2-meter telescope
time (shared with Dr. J. P. Huchra from
Harvard University) has been used to
obtain more photoelectric UBV photom-
etry for galaxies in the complete statisti-
cal sample of 1100 Zwicky galaxies, such
that I 6^1 I > 40, 8 > 0°, and m^w < 14.
Thuan plans to use the data: (1) to study
any systematic differences in properties,
such as color between cluster and non-
cluster galaxies; (2) to study the distri-

butions of color and absolute magnitude
of galaxies and to compare them with
simple models of star formation in gal-
axies; and (3) to provide independent
distance indicators (the color-magnitude
relation for early-type galaxies and the
H I width for the late-type galaxies) in
order to study grouping of galaxies, to
map the deviation from the Hubble flow,
and to deduce a local mean density.

The Palomar 1.5-meter telescope was
used to obtain surface photometry of the
inner parts of the same sample of gal-
axies. These data will be related to dy-
namical models.

Dr. Charles H. Townes of the Uni-
versity of California, Berkeley, in col-
laboration with F. Baas and J. Lacy
made observations in April of the 12.8
fi line of Ne II in the galactic center and
in a number of planetary nebulae. They
used the new 2.5-meter du Pont Tele-
scope of the Las Campanas Observatory
in Chile. The primary objective was to
obtain velocity and spatial distribution
of the ionized gas component of material
in the galactic center. Ne II provides
one of the most convenient probes into
this region. Spectra were taken in 24
fields of view near the galactic center,
each of diameter 7'' and with a spectral
resolution corresponding to a velocity
spread of about 40 km s~^. In addition,
the distribution of Ne within three nar-
row ranges of velocities was measured in
a number of spatial traces through the
region of the galactic center. The Doppler
velocities for these spatial distributions
were about +10, —160, and —240 km
s~^. Tracking controls of the telescope
worked very satisfactorily for both the
discrete fields of view and for the spatial
traces.

The spectrometer used was a tandem
Fabry-Perot grating system with com-
pletely cooled optics so that sensitivity
was limited primarily by photon noise
of radiation emitted in the path between
the spectrometer and the object observed,
and with the narrow bandwidth being
detected. Transparency of the telescope

182

CARNEGIE INSTITUTION

and atmosphere at Las Campanas was
good enough to reduce this noise fluctu-
ation substantially below the value ob-
tained when the spectrometer was look-
ing at room temperature radiation.

Radiation from the Ne II line was also
measured in a number of planetary
nebulae. However, most of the work was
directed toward exploring the galactic
center.

The spectra obtained from the galactic
center showed remarkable variations in
intensity of the Xe line and in its velocity
distribution between fields of view sepa-
rated on the relatively small scale of 5",
or about i,; pc. The majority of the Ne
is redshifted by amounts varying up to
several hundred kilometers per second.
However, there are also substantial
amounts of blueshifted Ne with equally
large Doppler shifts. In some regions,
several velocity peaks were found, indi-
cating several individual clouds within
a given field of view. Further work on
Ne II in this region is planned, as well
as somewhat similar examination of the
galactic center region in the infrared
spectra of other ions.

Drs. J. T. Trauger of the University
of Wisconsin, ]\Iadison, and M. E.
Mickelson of Denison University have
extended the original work on the meas-
urement of the HD abundance on Jupiter
(Trauger, Roesler, Carleton, and Traub,
Astrophys. J. [Lett.], 184, L137, 1973),
from which an HD/H2 and ultimately a
D H ratio insensitive to details in the
Jovian atmospheric structure can be de-
rived. Jupiter was observed during the
pfTiod 2-11 December 1976 at the Mount
Wilson 1.5-meter telescope with a 60-mm
PEP.SIOS spectrometer. The originally
observed HD (4, 0) PI feature was re-
measured to somewhat greater accuracy,
yielding a value of 0.28 ± 0.04 mA for
the integrated Jovian disk; and a search
was made for adflitional features (spe-
cifically the [4, 0] P2 and [5, 0] Rl
lines) to confirm the original HD identi-
fication. During the nights of 1-4 March
1977, the spectra of Saturn and Uranus

in the vicinity of the HD (4, 0) PI line
were observed with the 5-meter telescope
at Palomar Mountain with a 150 mm
PEPSI OS spectrometer. Laboratory stud-
ies in support of these observations have
been carried out with a PEPSIOS spec-
trometer in association with both a 2-
meter White cell and the Denison 22-
meter White cell, yielding accurate
absorption strengths for the HD (4, 0)
RO, PI, P2, and (5, 0) Rl lines, and
the first laboratory measurement of the
H2 (4, 0) SI line needed for comparison
in the computation of the D/H ratio.
The theoretical absorption strengths for
both the HD and H2 lines as used in
previous work were found to be signifi-
cantly in error. Background spectra in
the vicinity of the HD lines have also
been obtained for CH4 and NH3 for use
in data reduction. Preliminary analysis
of these astronomical and laboratory
data yield a Jovian D/H ratio near 4 X
10~^, somewhat greater than independ-
ently estimated values for the primitive
solar nebula. Similar results are obtained
for Saturn, while only an upper limit on
D/H is possible from the Uranian data,
owing to greater photon noise and diffi-
culties with atmospheric seeing and a
strong CH4 background. Refinements in
the analysis of the planetary and labora-
tory data are in progress; an improved
D/H ratio for Jupiter and the first esti-
mate of the D/H ratio on Saturn should
be forthcoming.

The two-stage image-tube spectro-
graph of the Palomar 1.5-meter telescope
was used by Dr. Barry E. Turnrose of
the NASA Johnson Space Center to ob-
tain calibrated photographic spectra of
a number of galaxies at dispersions of
140 and 60 A mm~^ Five members of
the relatively nearby compact cluster
Shakhbazian 30 were observed in the
blue spectral region at low dispersion for
the purpose of determining radial veloc-
ities. An extensive series of spectra at
position angles of 10°, 55°, 90°, and 135°
were obtained of the nucleus of the
peculiar spiral NGC 1614 (= Arp 186).

HALE OBSERVATORIES 183

These spectra were taken primarily at of Alabama conducted a search for com-

60 A mm~^, with emphasis on the spec- pact galaxies on long-exposure Illa-J

tral region near Ha. Their purpose is to Schmidt plates taken by Arp and Sulentic,

define more closely the internal kinemat- together with short-exposure plaU-s taken

ics of this unusual system. by Kowal and the late F. Zwicky. About

The object Pegasus 112, an apparent 200 compact or possibly compact galaxies

Seyfert galaxy discovered by G. Chin- brighter than 19th magnitude were cata-

carini, was observed in three separate logued in fields of 1° radius surrounding

spectral ranges between 3700 A and 7000 nine edge-on nearby spiral galaxies. Corn-

A at 60 A mm"^. The spectra clearly pact galaxies fainter than 18th magni-

delineate the profiles of the strong emis- tude show a marked anisotropy relative

sion lines whose blue wings exhibit widths to the major axis of the central spiral,

of several thousand km s~^. such that fewer compacts are found

Spectra of the galaxies NGC 1073, within 30° of the minor axis direction

NGC 1637, and NGC 3521 were also than in other directions (the deficiency

obtained, as were widened spectra of the is significant above the 3(7 level). The

spectrophotometric standard star HD result agrees with previous work about

84937 (for calibration purposes). distribution of brighter compacts near

Dr. P. D. Usher from the Pennsylvania edge-on spirals and indicates possible
State University was a guest investi- connection between the two classes,
gator during the summer 1976. He con- Dr. Sidney van den Bergh of the Uni-
tinued research on faint blue objects in versity of Toronto made use of the 5-
the Sandage-Luyten survey fields of the meter Hale Telescope. He photographed
North Galactic Polar Cap. Work com- the parent galaxy of the radio source
menced with the location of LB objects Cygnus A with various plate-filter com-
and their identification on finder charts, binations at the prime focus, showing
using the original three-color plates and that this object consists of a diffuse
related material. The objects were also source of continuous light on which a
identified and marked on a master 103a- huge boomerang-shaped emission region
plate. About 1200 objects of all three is superimposed. An astrometric investi-
color classes in both the 8-hour and 15- gation undertaken in collaboration with
hour fields were located and examined Kronberg and Button shows that the
for morphological characteristics. A central component of the radio source
master list relating LB and provisional Cygnus A lies halfway between the
numbers for the 15^ field was compiled, brightest part of the continuum source
A survey for completeness of objects with and the emission region. Photometric
U — V ^ 0.0 was completed in both observations and the fact that the radio
fields, and an additional list of faint blue source is located at the center of sym-
objects was compiled, along with their metry of the cD envelope of this galaxy
finding charts and estimated UBV color suggest that the true nucleus, in which
indices and magnitudes. A preliminary the central source of Cygnus A is em-
estimate of the degree of completeness bedded, is hidden from view by absorp-
was carried out for both fields. The posi- tion. Limiting Illa-J plates show that
tions of the supplementary objects and any optical object associated with the
standard stars in the 8^ field were meas- brighter radio components of the bright
ured on the X-Y machine in Pasadena, outer radio source must be fainter than
Measurements were reduced with the Mj ^ —14 + ^ log (H/lOO).
help of a program written by Kristian to Photographic and photometric obser-
give positions to 0.6" (rms) in both vations of two highly reddened galaxies
coordinates. that have recently been discovered in

Dr. Mauri Valtonen of the University Cygnus by Weinberger et al. show that

1S4 CARNEGIE INSTITUTION

these objects are not. as had previously were found to have strong Ha in emis-

been suspected, members of the Local sion. This is a much lower percentage

Group. than was anticipated. The presence (or

In coHaboration with deRoux, van den absence) of Ha in emission does not cor-

Bergh has analyzed the galactic stellar relate strongly with any other observable

population in a field surrounding the parameter. Follow-up photoelectric mon-

globular cluster XGC 6528. The redden- itoring of one of the M dwarfs with Ha

ing in this field is found to be somewhat in emission resulted in the detection of

higher than it is near XGC 6522. A two strong flares. This result follows the

fragmentary color-magnitude diagram of general correlation between the presence

the very compact cluster XGC 6528 itself of hydrogen lines in emission and flare

suggests that this object is quite metal activity. Dr. 0. Hansen agreed to monitor

rich. the star GL 569 and detected a strong

Proper-motion observations covering flare followed by a strong secondary

the period 1942-1976 (the earlier ma- flare.

terial having been obtained by Baade) Dr. George Wallerstein of the Uni-

show that the expansion time scale for versity of Washington has used the 2.5-

the optical remnant of SN 1604 is > meter coude with the Varo image tube

1 X ICH years. This indicates that the as well as the 5-meter coude spectro-

remnant consists of circumstellar ma- graph with the 90-mm image tube to

terial that was excited by the supernova investigate the abundances of the light

shell. The optical remnant is found to elements, especially oxygen, in stars that

have a tangential velocity of 525 it 117 show various composition anomalies,

km s~^ This suggests that Kepler's With the 2.5-meter, several stars with

supernova (which was of type 1) was abnormally strong or weak CH were

produced by a star with a Population observed in the near infrared for the I

Il-type orbit. This work was done by lines as well as for ON and NG in the

van den Bergh in collaboration with Dr. ultraviolet. The 5-meter was used to ob-

Karl Kamper. A few emission flocculi in serve extremely metal-poor stars for I

the remnant of the supernova have absorption at A6300. The variables SX

brightened significantly during the last Herculis, TY Virginis, and CK Virginis

35 years. X^o flocculi, however, have faded were included. The variable carbon stars

during the same period. According to AC Herculis and V553 Centauri were

van den Bergh, color observations as well also observed. All the spectrograms have

as photometry of field stars give a fore- been traced and are being measured

ground reddening Eb - v =^ 0.7 ± 0.2. during the summer of 1977.

This reddening value and the assumption Wallerstein continued to monitor two

that SX' 1604 occurred in the nuclear maser stars: CY Canis Majoris in Oc-

bulge of the Galaxy yield Mv (max) = tober 1976 and VX Sagittarii in October

-19.3. 1976 and April 1977.

Dr. G. Veeder of the Jet Propulsion Dr. Richard M. West of the European
Laboratory has initiated a photometric Southern Observatory carried out spec-
and ."spectral survey of intrinsically faint troscopic observations of southern gal-
M dwarf.s. This program concentrates on axies during observation runs at Las
identifying late M dwarfs that show Campanas in October 1976 and March
chromospheric activity either by hydro- 1977. He used the 1-meter Swope Tele-
gen lines in emission or by flares. Ap- scope and image-tube spectrograph. Ap-
proximately 30 candidate stars have been proximately 200 galaxies were observed,
included in the program. Image-tube most at dispersion 284 A mm~^ but
spectra (~ 144 A mm~^) have been several also at 135 A mm~^. A majority
obtained for 15 stars. Three candidates of the galaxies were recent discoveries

HALE OBSERVATORIES

185

on ESO Schmidt plates. About half
showed emission lines, and several new
Seyfert galaxies were identified. A few
dubious objects could be confirmed as
planetary nebula. The results are being
published in Astronomy and Astrophysics.

As a result of this continuing investiga-
tion, West expects to carry out a detailed
astrophysical study of the galaxies with
particularly active nuclei by means of
high-dispersion spectra from the ESO
3.6-meter telescope.

BIBLIOGRAPHY

Aaronson, Marc, see Frogel, Jay A.; Persson,
S. Eric.

Adams, W. M., Differential rotation of pho-
tospheric magnetic fields associated with
coronal holes, Sol Phys., 47, 601-605, 1976.

Allen, D. A., see Boksenberg, A.

Arp, Halton, Ejection from the spiral galaxy
NGC 1097. Astrophys. J, (Lett.), 207, L147-
L150, 1976.

Arp, Halton, A quasar near a companion gal-
axy, NGC 5296. Astrophys. J. (Lett.), 210,
L59-L61, 1976.

Arp, Halton, Photography of faint surface
brightness features in galaxies, in Proc, lAU
Working Group on Photographic Problems
(Commission 9), pp.- 83-84, J. L. Heudier,
ed., Observatoire de Nice, 1976.

Arp, Halton, Anomalous redshifts in galaxies
and quasars, in Redshifts and the Expansion
of the Universe, Int. Astron. Union Colloq.,
37-Centre Nat. Recherche Sci. Colloq., 263,
pp. 377-398, C. Balkowski and B. E. Wester-
lund, eds.. Centre National Recherche Scien-
tifique, Paris, 1976.

Arp, Halton, and Jean Lorre, Image processing
of galaxy photographs, Astrophys. Jr., 210,
58-64, 1976.

Arp, Halton, and Barry F. Madore, Preliminary
results from the catalogue of southern peculiar
galaxies and associations, Quart. J. Roy.
Astron. Soc, 18, 234-241, 1977.

Arp, Halton, Jack W. Sulentic, A. G. Willis,
and H. R. de Ruiter, A BL Lac object near
the spiral galaxy NGC 6503, Astrophys. J.
(Lett.), 207, L13-L16, 1976.

Arp, Halton, see also de Ruiter, H. R.; Kirsh-
ner, Robert P.

Barbieri, C, and Jack W. Sulentic, The plane-
tary nebula 146+31 ?1, Publ. Astron. Soc.
Pacific, 89, 261-263, 1977.

Baschek, B., and Anneila I. Sargent, The chem-
ical composition of the horizontal-branch star
Feige 86, Astron. Astrophys., 63, 47-52, 1976.

Becklin, Eric E., K. Matthews, G. Neugebauer,
and C. G. Wynn-WiUiams, On the infrared
emission from Sgr B2, Astron. Astrophys., 55,
19-21, 1977.

Becklin, Eric E., S. Beckwith, I. Gatley, K.
Matthews, G. Neugebauer, C. Sarazin, and
M. W. Werner, Infrared studies of an ioniza-

tion front region in the Orion Nebula,
Astrophys. J., 207, 770-779, 1976.

Becklin, Eric E., see also Beckwith, S.; Ennis,
D.; Frogel, Jay A.; Mason, K. 0.; Wynn-
Wilhams, C. G.

Beckwith, S., N. J. Evans II, E. E. Becklin,
and G. Neugebauer, Infrared observations of
Mon R2, Astrophys. J., 208, 390-395, 1976.

Beckwith, S., see also Becklin, Eric E. ; Ennis, D.

Berghstralh, Jay T., see Miinch, Guido.

Blankenship, L., see Mason, K. O.

Boksenberg, A., S. A. Briggs, R. F. Carswell,
M. Schmidt, and D. Walsh, 2005+403— A
QSO near the galactic plane, Mon. Notic.
Roy. Astron. Soc, 177, 43P-45P, 1976.

Boksenberg, A., R. F. Carswell, D. A. Allen,
R.A.E. Fosbury, M. V. Penston, and W.L.W.
Sargent, The remarkable Seyfert galaxj'
Markarian 231, Mon. Notic. Roy. Astron.
Soc, 178, 451^66, 1977.

Boksenberg, A., see also Green, Richard F.;
Greenstein, Jesse L.; Sargent, Wallace L. W.

Bond, Howard E., see Green, Richard F.

Boroson, T. A., see Sargent, Wallace L. W.

Borra, Ermanno F., and Arthur H. Vaughan.
Observations of the transverse Zeeman effect
in the magnetic star /3 Coronae Boreahs:
evidence for the oblique rotator model,
Astrophys. J. (Lett.), 210, L145-L147, 1976.

Briggs, S. A., see Boksenberg, A.

Brown, R, L., see Mason, K. 0.

Campbell, M. F., J. H. Ehas, D. Y. Gezari,
P. M. Harvey, W. F. Hoffman, H. S. Hudson,
G. Neugebauer, B. T. Soifer, M. W. Werner,
and W. E. Westbrook, Far infrared observa-
tions of IRC+ 10216, Astrophys. J.. 208. 396-
398, 1976.

Carswell, R. F., see Boksenberg, A.; Greenstein,
Jesse L.

V

Chase, R. C, see Svestka, Z.

Chase, S. C, Jr., E. D. Miner, D. Morrison,
G. Miinch, and G. Neugebauer, Mariner 10
infrared radiometer results: temperatures and
thermal properties of the surface of Mercury,
Icarus, 28, 565-578, 1976.

Chase, S. C, Jr., see also Kieffer, H. H.

Dahn, Conrad C, see Zinn. Robert.

de Bruyn, A. G., see Segalovitz, A.

Dennison, E. W., Memory systems for signal
generating photoelectric image detectors, in

186

CARNEGIE INSTITUTION

Adixinces in Elect rati ics and Electron Physics,
Vol. 40B. Photo-Electronic Image Devices,
pp. 72\>-733. B. L. Morgan. R. W. Airoy. and

D. McMuIlan. eds.. Academic Pres8, London,
1976.

de Riiiter. H. R.. A. G. Willis, and H. C. Arp,
A Westerbork 1415 MHz survey of back-
ground radio sources. II. Optical identifica-
tions with deep Illa-J plates. Aslron. Astro-
phys.. Suppl. Ser.. -JS, 211-277, 1977.

de Ruiter. H. R.. .^rr also Arp. Halt on.

Diner. D. J.. J. A. Westphal. F. P. Schloerb,
Infrared imaging of Venus: 8-14 micrometers,
Icarus. 27, 191-195. 1976.

Dunlap, J. R.. see Kirshner. Robert P.

Elia.<. J. H., see Campbell. M. F.; Ennis. D.;
Mason. K. O.; Westbrook. W. E.

Ennis. D.. E. E. Becklin. S. Beckwith, J. Elias,
I. Gatley. K. Matthews, G. Neugebauer, and
P. Willner. Infrared observations of Nova
Cygni 1975. Astrophys. J.. 21If, 478-487, 1977.

Etzel. Paul B., see Lanning, Howard.

Evans. X. J.. II, see Beckwith. S.

Forster. J. R.. see Wynn-WiUiams, C. G.

Fosbury. R. A. E.. see Boksenberg, A.

Frogel. Jay A., S. Eric Persson, and Marc
Aaronson. Compact infrared sources associ-
ated with southern H II regions, II, Astro-
phys. J., 213, 723-736, 1977.

Frogel, Jay A., S. E. Persson, M. Aaronson,

E. E. Becklin. and K. Matthews, Infrared
observations of the late-type stellar com-
ponent of elliptical galaxies, in The Galaxy
and the Local Group (RGO Bull. No. 182),
pp. 111-118, R. J. Dickens and J. E. Perry,
eds.. Royal Greenwich Observatory, Herst-
monceux. England, 1976.

Frogel, Jay A., see also Persson, S. Eric.

Gatley. 1., see Becklin, E. E.; Ennis, D.

Gezari. D. Y., see Campbell. M. F.; Westbrook,
W. E.

Goldreich, Peter, and Wallace L. W^. Sargent,
Qua.'sar absorption lines, Comments Astrophys.
Space Phys., 6, 133-137, 1976.

Gott, J. Richard, III, and Edwin L. Turner,
The mean luminosity and mass densities
in the univer.s«, Astrophys. J., 209, 1-5, 1976.

Gott, J. Richard, III, see also Turner, Edwin L.

Green. Richard F., A discover>^ program for
bright qua.'^ars: preliminary results, Publ.
Astron. Soc. Pacific, 88, 66.5^668, 1976.

Green, Richard F., and Douglas 0. Richstone,
On the reality of periodicities in the redshift
di.'stribution of emission-line objects, Astro-
phys. J.. 208, 639-645, 1976.

Green, Richard F., Jesse L. Greenstein, and A.
Boksenbr-rg. A blue variable at high latitude,
Publ. Astron. Soc. Pacific, 88, 598-^02, 1976.

Green. Richard F., John P. Huchra, and Howard
E. Bond, A study of the optical variability
of the Seyfert galaxy X Comae, Publ. Astron.
Soc. Pacific, 89, 2.5.5-260, 1977.

Greenstein, Jesse L., Some further degenerate
stars, IX, Astrophys. J. {Lett.), 207, L119-
L123, 1976.

Greenstein, Jesse L., Degenerate stars with
helium atmospheres, Astrophys. J., 210, 524-
532, 1976.

Greenstein, Jesse L., Subluminous stars: post-
novae, white dwarfs, and hot subdwarfs,
Mem. Soc. Roy. Scis. Liege, 9, 247-267, 1976.

Greenstein, Jesse L., Bond's flare star 232^
03, Publ. Astron. Soc. Pacific, 89, 304-308,
1977.

Greenstein, Jesse L., and J. B. Oke, An interpre-
tation of the spectrum of the red rectangle,
Publ. Astron. Soc. Pacific, 89, 131-138, 1976.

Greenstein, Jesse L., M. A. Kazarian, and E.
Ye. Khachikian, On the electron density of
cometary nebula NGC 2261, Pisma in
Astronomy, 1, 26-27, 1975.

Greenstein, Jesse L., A. Boksenberg, R. Cars-
well, and K. Shortridge, The rotation and
gravitational redshift of white dwarfs,
Astrophys. P., 212, 186-197, 1977.

Greenstein, Jesse L., M. A. Kazarian, T. Yu.
Magakian, and E. Ye. Khachikian, A spectro-
photometry of NGC 2261 and R. Mon, I,
Astrofizika, 12, 587-611, 1976.

Greenstein, Jesse L., see also Green, Richard
F.; Oke, J. B.

Griffin, R. F., and James E. Gunn, Hyades
giants d and 6^ Tauri as spectroscopic binaries,
Astron. J., 82, 176-178, 1977.

Gunn, James E., Observational tests in cos-
mology, in Redshijts and the Expansion of
the Universe, Int. Astron. Union Colloq.,
37-Centre Nat. Recherche Sci. Colloq., 263,
pp. 183-206, C. Balkowski and B. E. Wester-
lund, eds.. Centre National Recherche Scien-
tifique, Paris, 1976.

Gunn, James E., and Beatrice M. Tinsley,
Dynamical friction: the Hubble diagram as
a cosmological test, Astrophys. J., 210, 1-6,
1976.

Gunn, James E., see also Griffin, R. F.; Steig-
man, Gary; Thuan, Trinh X.

Hardy, Eduardo, The population structure of
the LMC disk, I, Photographic and photo-
electric observations of the underlying stellar
component of the central regions, Astrophys.
J., 211, 718-727, 1977.

Hartoog, Mark R., The rotation of Ap stars
in open clusters and magnetic braking,
Astrophys. J., 212, 723-727, 1977.

Harvey, P. M., see Campbell, M. F.

Hauser, M. G., see Werner, M. W.; Westbrook,
W. E.

Hickson, Paul, Douglas 0. Richstone, and
Edwin L. Turner, Galaxy collisions in dense
groups, Astrophys. J., 213, 323-326, 1977.

Hjellming, R. M., see Mason, K. O.

Hoessel, John, see Melnick, Jorge.

Hoffmann, W. F., .see Campbell, M. F.

HALE OBSERVATORIES

187

Houck, J. R., see Werner, M. W.

Howard, Robert, A possible variation of the
solar rotation with the activity cycle, Astro-
phys. J. (Lett.), 210, L159-L161, 1976.

Howard, Robert, Phenomenological model of
the solar cycle, in Physics of Solar Planetary
Environments, Vol. 1, pp. 34^3, Donald J.
Williams, ed., American Geophysical Union,
Boulder, Colorado, 1976.

Howard, Robert, Studies of solar magnetic
fields, IV, The effects of angular resolution,
Sol. Phys., 47, 57&-580, 1976.

Howard, Robert, The Mount Wilson solar
magnetograph scanning and data system, Sol.
Phys., 48, 411-416, 1977.

Howard, Robert, Studies of solar magnetic
fields, V, The true average field strengths
near the poles, Sol. Phys., 52, 243-248, 1977.

Howard, Robert, Non-spot magnetic fields, in
Transactions of the International Astronomi-
cal Union, Reports on Astronomy, Vol,
XVI-A, Part 2, pp. 41-43, G. Contopoulos,
ed., D. Reidel Publishing Co., Dordrecht-
Holland, 1976.

Howard, Robert, and Z. Svestka, Development
of a complex activity in the solar corona,
Sol. Phys., in press, 1977.

Howard, Robert, and Hirokazu Yoshimura,
Differential rotation and global-scale velocity
fields, in Basic Mechanisms of Solar Activity,
Int. Astron. Union Symp., 71, pp. 19-35, V.
Bumba and J. Kleczec, eds., D. Reidel Pub-
lishing Co., Dordrecht-Holland, 1976.

Hmvard, Robert, see also Scherrer, Philip H.;
Svestka, Z.

Huchra, John, The Zwicky catalogue magnitude
scale : a comparison with photoelectric mag-
nitudes of faint galaxies and with the magni-
tudes of Holmberg and de Vaucouleurs,
Astron. J., 81, 952-953, 1976.

Huchra, John. Comment on photometric evo-
lutionary corrections, in Redshifts and the
Expansion of the Universe, Int. Astron. Union
Colloq., 37-Centre Nat. Recherche Sci.
Colloq., 263, pp. 179-181, C. Balkowski and
B. E. Westerlund, eds.. Centre National
Recherche Scientifique, Paris, 1976.

Huchra, John P., The mass-to-light ratio of
Markarian galaxies, in Redshifts and the
Expansion of the Universe, Int. Astron. Union
Colloq., 37-Centre Nat. Recherche Sci.
Colloq., 263, pp. 289-299, C. Balkowski and
B. E. Westerlund, eds.. Centre National
Recherche Scientifique, Paris, 1976.

Huchra, John P., see also Green, Richard F.;
Kowal, C.

Hudson, H. S., see Campbell, M. F.

Jensen, E. B., see Strom, K. M.

Johnson, Harold L., see Sandage, Allan.

Katem, Basil, see Sandage, Allan.

Kazarian, M. A., see Greenstein, Jesse L.

Keenan, Phihp C, and Olin C. Wilson, Effects

of heavy-element abundance on .spectro-
scopic luminosities of G5 KO giants, Astro-
phys. J., 212, 39f>-407, 1977.

Khachikian, E. Ye., see Greenstein, Je.sse L.

Kieffer, H. H., Stillman C. Chase, Jr., Ellis D.
Miner, Frank Don ]-*alluconi, Guido Miinch,
Gerry Neugebauer, and Terry Z. Martin,
Infrared thermal mapping of the martian
surface and atmosphere : first results, Science,
193, 780-786, 1976.

Kirshner, Robert P., and Keith Taylor, High-
velocity gas in the Cygnus Loop, Astrophys.
J. (Lett.), 208, L83-L86, 1976.

Kirshner, Robert P., H. C. Arp, and J. R.
Dunlap, Observations of supernovae: 1975a
in NGC 2207 and 1975b in the Per.seus
cluster, Astrophys. J., 207, 44-52, 1976.

Kormendy, John, Brightness distributions in
compact and normal galaxies, I, Surface pho-
tometry of red compact galaxies, Astrophys.
J., 214, 359-382, 1977.

Kormendy, John, Absence of velocity anisot-
ropy in the direction of the Virgo cluster, in
Redshifts and the Expansion of the Universe,
Int. Astron. Union Colloq., 37-Centre Nat.
Recherche Sci. Colloq., 263, pp. 155-158, C.
Balkowski and B. E. Westerlund, eds.. Centre
National Recherche Scientifique, Paris, 1976.

Kowal, Charles T., The case against names,
Icarus, 29, 513, 1976.

Kowal, Charles T., Periodic Comet Taylor
(1916 I = 1977a), Int. Astron. Union Circ,
3033, January 28, 1977.

Kowal, Charles T., 1936CA (Adonis), Int.
Astron. Union Circ, 3041, February 23, 1977.

Kowal, Charles T., 1977HB, Int. Astron. Union
Circ, 3066, April 26, 1977.

Kowal, Charles T., Comet Kowal (1977f), Int.
Astron. Union Circ, 3066, April 26, 1977;
3067, April 29, 1977; 3076, May 20, 1977;
3079, May 31, 1977.

Kowal, Charles T., J. Huchra, and W. L. W.
Sargent, The 1975 Palomar supernova search,
Publ. Astron. Soc Pacific, 88, 521-525, 1976.

Krieger, A. E., see Svestka, Z.

Kristian, Jerome, The present status of 3CR
identifications, in Radio Astronomy and Cos-
mology, Int. Astron. Union Symp., 74, D. L.
Jauncey, ed., D. Reidel Pubhshing Co.,
Dordrecht-Holland, 1977.

Kristian, Jerome, A. R. Sandage, and J. A.
Westphal, The Hubble diagram for red mag-
nitudes of bright cluster galaxies, in Redshifts
and the Expansion of the Universe, Int.
Astron. Union Colloq., 37-Centre Nat. Re-
cherche Sci. Colloq., 263, pp. 309-311, C.
Balkowski and B. E. Westerlund, eds.. Centre
National Recherche Scientifique, Paris, 1976.

Kristian, Jerome, see also Westphal, James A.

LaBonte, Barry J., Ha observations of the
August 12, 1975, type III-RS bursts, Sol.
Phys., 50, 201-211, 1976.

188

CARNEGIE INSTITUTION

LaBonte. Bvirn- J.. A measurement of the
helium D3 protiK^ with :\ birefringent filter.
Sol. Phys., in press. 1977.

Lanning. Howard, and Paul B. Etzel. Ha emis-
sion in the eclipsing white dwarf V471 Tau
(BD-}-16'516). ComrnL<!s{on 37 Int. Astron.
I'nion Bull. Var. Stars. 1147, 29 June 1976.

Lo. K. Y.. sec Westbrook. W. E.

Lorre. Jean, see Arp. Halton.

Madore. Barry F., see Arp. Halton.

Magakian. T. Yu., sec Greenstein, Jesse L.

Marsh. K. A.. The lifetimes and evolution of
fibrils, Sol. Phys.. 60. 37-48. 1976.

Marsh. K. A., The calcium K-line network in
coronal holes. Sol. Phys., in press, 1977.

Martin, Terry Z., see Kieffer, H. H.

Mason. K. 0.. E. E. Becklin, L. Blankenship,
R. L. Brown. J. Ehas. R. M. HjcUming. K.
Matthews. P. G. Murdin. G. Xeugebauer, P.
W. Sanford. and S. P. Willner, Further joint
x-rav. infrared, and radio observations of
Cygnus X-3. Asfrophys. J.. 207, 78-87, 1976.

Massey. Philip. Photometiy of Shakhbazian I,
Publ. Astron. Soc. Pacific, 89, 13-18, 1977.

Matthews. Keith, see Becklin, Eric E.; Ennis,
D.: Frogel. Jav A.; Mason, K. O.; Wynn-
WilHams. C. G.

Melnick. Jorge, Velocity dispersions in giant
H II regions: relation with their linear
diameters, Astrophys. J.. 213, 15-17, 1977.

Melnick. Jorge, Simon D. M. White, and John
Hoe.ssel. Photoelectric surface photometry of
the Coma Cluster. Man. Notic. Roy. Astron.
Soc. ISO, 207-218, 1977.

Miner, E. D.. see S. C. Chase, Jr.; KiefTer,
H.H.

Moller. J., see Strom, K. M.

Moore, R. L., see Zirin, Harold.

Morrison, D., see Cha.«e, S. C, Jr.

Miinch. Guido, Herbig-Haro object in the Orion
Nebula, Astrophys. J. (Lett.), 212, L77-L79,
1977.

Mijnch, Guido, Highlights in planetary spec-
troscopy 1962-75, Mem. Soc. Roy. Scis. Liege,
9, 87-100. 1976.

Miinch, Guido, and Jay T. Bergstralh, lo:
morphologj' of its sodium emission region,
Publ. A.stion. Soc. Pacific, 89, 232-237, 1977.

Miinch, Guido, see also Chase, S. C, Jr.;
Kif'ffer, H. H.

Murdin, P. G.. see Mason, K. 0.

Xeugebauer, Gerry, see Becklin, Eric E.; Beck-
with, S.; Campbell, M. F.; Chase, S. C, Jr.;
Enni.s, D.; Kieffer, H. H.; Mason, K. 0.;
Wf-mer, M. W.; We.stbrook, W. E.; Wynn-
Williams, C. G.

Oke, J. B., Spectrophotometry of the x-ray
binar>' HZ Herculis, Astrophys. J., 209, 547-
555, 1976.

Oke, J. B., and J. L. Greenstein, Spectro-
photometr>' of the transient x-ray source
A0620-00, Astrophys. .J., 211, 872-880, 1977.

Oke, J. B., and G. A. Shields, Seyfert galaxies
with strong Fe II emission, Astrophys. J.,
207, 713-724, 1976.

Oke, J. B., see also Greenstein, Jesse L.; Rich-
stone. D. O.

Palluconi, Frank D., sec Kieffer, H. H.

Penston, M. V., see Boksenberg, A.

Persson, S. Eric, Jay A. Frogel, and Marc
Aaronson, The 10 micron silicate feature in
southern H II regions, Astrophys. J., 208,
753-759, 1976.

Persson, S. Eric, see also Frogel, Jay A.

Richstone, D. 0., and J. B. Oke, The nebulosity
near the quasar 3C249.1, Astrophys. J., 21S,
8-14, 1977.

Richstone, D. O., see also, Green, Richard E.;
Hickson, Paul; Tremaine, Scott D.

Sandage, Allan, High-latitude reflection nebu-
losities illuminated by the galactic plane,
Astron. J., 81, 954-957, 1976.

Sandage, Allan, The reddening and extinction
of a cluster of bright galaxies near the galactic
plane in Cygnus, Publ. Astron. Soc. Pacific,
88, 367-369, 1976.

Sandage, Allan, and Basil Katem, The brightest
stars in nearby galaxies, I, The color-magni-
tude diagram and luminosity function for
IC 1613, Astron. J., 81, 743-750, 1976.

Sandage, Allan, and G. A. Tammann, The
Virgo cluster, I, The equality of mean red-
shifts of E and S galaxies near the cluster
center, Astrophys. J. (Lett.), 207, L1-L4, 1976.

Sandage, Allan, and G. A. Tammann, Steps
toward the Hubble constant, VII, Distances
to NGC 2403, MlOl, and the Virgo cluster
using 21-cm line widths compared with opti-
cal methods: the global value of Ho, Astro-
phys. J., 210, 7-24, 1976.

Sandage, Allan, Basil Katem, and Harold L.
Johnson, The halo globular cluster NGC
5053, Astron. J., 82, 389-394, 1977.

Sandage, Allan, see also Kristian, Jerome.

Sanford, P. W., see Mason, K. 0.

Sarazin, C, see Becklin, Eric E.

Sargent, Anneila I., see Baschek, B.

Sargent, Wallace L. W., Redshifts for six 3CR
radio galaxies and the spectrum of 3C111,
Astrophys. J. (Lett.), 212, L105-L106, 1977.

Sargent, Wallace L. W., Quasar absorption lines
as probes of the past intergalactic medium,
in The Evolution of Galaxies and Stellar
Populations, Proc. Yale University Confer-
ence, pp. 427-442, B. M. Tinsley and R. B.
Larson, eds., D. C. Gehret, asst. ed., Yale
University Observatory, New Haven, Con-
necticut, 1977.

Sargent, Wallace L. W., Radio sources and
clusters of galaxies in The Physics of Non-
Thermal Radio Sources, pp. 67-81, G. Setti,
ed., D. Reidel Publishing Co., Dordecht-
Holland, 1976.

Sargent, Wallace L. W., and T. A. Boroson,

HALE OBSERVATORIES

189

A statistical assessment of the evidence for
line-locking in quasar spectra, Astrophys. J.,
212, 383-389, 1977.

Sargent, Wallace L. W., and Edwin L. Turner,
A statistical method for determining the
cosmological density parameter from the red-
shifts of a complete sample of galaxies,
Astrophys. J. {Lett.), 212, L3-L7, 1977.

Sargent, Wallace L. W., Paul L. Schechter, A.
Boksenberg, and Keith Shortridge, Velocity-
dispersions for 13 galaxies, Astrophys. J.,
212 326-334, 1977.

Sargent, Wallace L. W., see also Boksenberg,
A.; Goldreich, Peter; Kowal, Charles T.

Schechter, Paul L., see Sargent, Wallace L. W.

Scherrer, PhiHp H., A. B. Severny, John M.
Wilcox, and Robert Howard, The mean mag-
netic field of the sun : method of observation
and relation to the interplanetary magnetic
field, Sol. Phys., 62, 3-12, 1977.

Schloerb, F. P., see Diner, D. J.

Schmidt, Maarten, On the apparent absence
of evolution of quasi-stellar radio sources
with fiat radio spectra, Astrophys. J. (Lett.),
209, L55-L57, 1976.

Schmidt, Maarten, Cosmological interpretation
of redshift data on quasars through the
V/Fmax test, in Radio Astronomy and Cos-
mology, Int. Astron. Union Symp. 74, in press,
D. L. Jauncey, ed., D. Reidel Publishing
Co., Dordrecht-Holland, 1977.

Schmidt, Maarten, see also Boksenberg, A.

Schramm, David N., see Steigman, Gary.

Schweizer, Francois, On the rotation and rela-
tive mass of NGC 5195, the companion of
M51, Astrophys. J., 211, 324-328, 1977.

Schweizer, Francois, Photometric studies of
spiral structure, I, The disks and arms of
six Sb I and Sc I galaxies, Astrophys. J.,
Suppl. Ser., 31, 313-332, 1976.

Searle, Leonard, Composition differences within
and among galaxies, in The Galaxy and the
Local Group (RGO Bull. No. 182), pp. 119-
127, R. J. Dickens and J. E. Perry, eds.. Royal
Greenwich Observatory, Herstmonceux, Eng-
land, 1976.

Searle, Leonard, The evolution of stellar popu-
lations, in The Evolution of Galaxies and
Stellar Populations, Proc. Yale University
Conference, pp. 219-237, B. M. Tinsley and
R. B. Larson, eds., D. C. Gehret, asst. ed.,
Yale University Observatory, New Haven,
Connecticut, 1977.

Searle, Leonard, see also Zinn, Robert.

Segalovitz, A., W. W. Shane, and A. G. de
Bruyn, Polarisation detection at radio wave-
lengths in three spiral galaxies. Nature, 264,
222-226, 1976.

Severny, A. B., see Scherrer, Philip H.

Shane, W. W., see Segalovitz, A.

Shields, G. A., see Oke, J. B.

Shortridge, K., see Greenstein, Jesse L,; Sar-
gent, Wallace L. W.

Soifer, B. T., see Campbell, M. F.

Steigman, Gary, David X. Schramm, and James
E. Gunn, Cosmological limits to the nurnbf;r
of massive leptons, Phys. Lett., OOB, 202-204,
1977.

Strom, K. M., S. E. Strom, E. B. Jensen, J.
Moller, L. A. Thompson, and Trinh X.
Thuan, A photometric study of the SO galaxy
NGC 3115, Astrophys. J., 212, 335-346, 1977.

Strom, S. E., see Strom, K. M.

Sulentic, Jack W., Radio emission and optical
morphology in Markarian galaxies, Astron. J.,
81, 582-590, 1976.

Sulentic, Jack W., Redshifts in the Virgo
cluster, Astrophys. J. (Lett.), 211, L59-L62,
1977.

Sulentic, Jack W., 1976WA, Int. Astron. Union
Circ, 3071, May 10, 1977.

Sulentic, Jack W., see also Arp, Halton;
Barbieri, C.

Svestka, Z., A. S. Krieger, R. C. Chase, and R.
Howard, Transequatorial loops interconnect-
ing McMath regions 12472 and 12474, Sol

^ Phys., 52, 69-90, 1977.

Svestka, Z., see also Howard, Robert.

Tammann, G. A., Galactic and extragalactic
supernova frequencies, in Proceedings of the
1976 Duman Summer Workshop, pp. 137-
162, Arthur Roberts, ed., CERN, Geneva,
Switzerland, 1976.

Tammann, G. A., A progress report on super-
nova statistics in Supemovae, Astrophysics
and Space Science Library, Vol. 6, pp. 95-116,
David N. Schramm, ed., D. Reidel Publish-
ing Co., Dordrecht-Holland, 1977.

Tammann, G. A., The Hubble constant and
the local expansion field, in Redshifts and the
Expansion of the Universe, Int. Astron. Union
Colloq., 37-Centre Nat. Recherche Sci.
Colloq., 263, pp. 43-65, C. Balkowski and
B. E. Westerland, eds.. Centre National
Recherche Scientifique, Paris, 1977.

Tammann, G. A., see also Sandage, Allan,

Tanaka, K., see Zirin, Harold.

Taylor, Keith, see Kirshner, Robert P.

Terrile, R. J., and J. A. Westphal, The vertical
cloud structure of Jupiter from 5 ixvo. meas-
urements, Icarus, 30, 274-281, 1977.

Terrile, R. J., and J. A. Westphal. Infrared
imaging of Jupiter in the 8-14 micrometer
spectral region, Icarus, 30, 730-735, 1977.

Thompson, L. A., see Strom, K. M.

Thuan, Trinh X., and James E. Gunn, A new
four-color intermediate-band photometric sys-
tem, Publ. Astron. Soc. Pacific. 88. 543-547,
1976.

Thuan, Trinh X., see also Strom, K. M.

Tinsley, Beatrice M., see Gunn. James E.

Tremaine, Scott D., and Douglas 0. Richstone,
A test of a statistical model for the luminosi-

190

CARNEGIE INSTITUTION

ties of bright cluster galaxies, Astrophys. J.,
212, 311^16, 1977.

Turner. Edwin L.. Binary galaxies, I, A well-
lietined statistical s^\mple. Astrophys. J., '^OS,
20-2V), 1976.

Turner. Edwni L., Binary galaxies, II, Dy-
namics and mass-to-light ratios. Astrophys. J..
20S, 304-316. 1976.

Turner. Edwin L.. and J. Richard Gott, III,
Groups of galaxies, I. A catalog, Astrophys.
J.. Suppl. Scr.. 32, 409-427. 1976.

Turner, Edwin L.. and J. Richard Gott, III,
Groups of galaxies. II. The luminosity func-
tion, Astrophys. J.. 309. 6-11. 1976.

Turner, Edwin L., .<rc also Gott, J. Richard,
III: Hickson. Paul; Sargent, Wallace L. W.

van der Kmit. P. C. Observations of velocity
tields in three spiral galaxies using optical
emission lines, Astron. Astrophys., Suppl. Ser.,
■25. 527-54S. 1976.

Vaughan. Artluir H.. see Borra, Ermanno F.

Wade. Richard A.. Supernova in NGC 5427,
Int. Astron. Union Circ, 29S4, August 30,
1976.

Walsh. D., see Boksenberg, A.

Walter, J., see Zirin. Harold.

Welch. W. J., see Wynn-Wilhams, C. G.

Werner, M. W., G. Xeugebauer, J. R. Houck,
and M. G. Hauser, One-millimeter observa-
tions of the Crab Nebula. Publ. Astron. Soc.
Pacific. S9, 127-128. 1977.

Werner. M. W., see also, Beckhn, Eric E.;
Campbell. M. F.; Westbrook, W. E.; Wvnn-
Wilhams, C. G.

Westbrook, W. E., M. W. Werner, J. H. Elias,
D. Y. Gezari, M. G. Hauser, K. Y. Lo, and
G. Xeugebauer, One-millimeter continuum
emission studies of four molecular clouds,
Astrophys. J., 209, 94-101, 1976.

Westbrook, W. E., .see also Campbell, M. F.

Westphal, James A., and Jerome Kristian, The
use of silicon target (SIT) camera tubes for
photometry, spectroscopy, and field viewing,
in Astronomical Application of Image De-
tectors and Linear Response., Int. Astron.
Union Colloq., 40, pp. 19-1-19-21, M.
Duchesne and G. Lelievre, eds., Obsorvatoire
de Paris-Meudon. France, 1976.

We.stphal, James A., see also Diner, D. J.;
Kristian, Jerome; Terrile, R. J.

White, Simon D. M., see Melnick, Jorge,

Wilcox, John M., see Scherrer, Philip H.

Williams, T. B., Population synthesis of the
nnr-lpi of ten nearby galaxies, Astrophys. J.,
209, 716-733, 1976. '

Williams, T. B., Velocity dispersions in the
nufloi of Tf-n nearbv galaxies, Astrophys. J.,
21/,, 685-691, 1977.

Willis, A. G., see Arp, Halton; de Ruiter, H. R.

Willncr, S. P., 8 to 13 micron spectrophotometry
of compact sources in W3, Astrophys. J., 214,
706-711, 1977.

WiUner. S. P., see also Ennis D.; Mason, K. O.

Wilson, Olin C, Chromospheric variations in
main-sequence stars, in Basic Mechanisms of
Solar Activity, Int. Astron. Union Symp., 71,
pp. 447-452, Y. Bumba and J. Kleczec, eds.,
D. Reidel Publishing Co., Dordrecht-Holland,
1976.

Wilson, Olin C, see also Keenan, Philip C.

Wright, M. C. A., see Wynn-Williams, C. G.

Wynn-Williams, C. G., E. E. Becklin, J. R.
Forster, K. Matthews, G. Neugebauer, W. J.
Welch, and M. C. A. Wright, On the rela-
tionship between the infrared source CRL
2591 (UOA-27) and its radio and H2O coun-
terparts, Astrophys. J. (Lett.), 211, L89-L90,
1977.

Wynn-Wilhams, C. G., E. E. Beckhn, K.
Matthews, G. Neugebauer, and M. W. Wer-
ner, Infrared studies of H II regions and
dust clouds near K3-50, Mon. Notic. Roy.
Astron. Soc, 179, 255-264, 1977.

Wynn-Williams, C. G., see also Becklin, Eric E.

Yoshimura, Hirokazu, Solar cycle evolution of
the general magnetic field, in Basic Mecha-
nisms of Solar Activity, Int. Astron. Union
Symp, 71, pp. 137-143, V. Bumba and J.
Kleczec, eds., D. Reidel Publishing Co.,
Dordrecht-Holland, 1976.

Yoshimura, Hirokazu, A model of the solar
cycle driven by the dynamo action of the
global convection in the solar convection
zone, in Basic Mechanisms of Solar Activity,
Int. Astron. Unioyi Symp., 71, pp. 409^13,
V. Bumba and J. Kleczec, eds., D. Reidel
Publishing Co., Dordrecht-Holland, 1976.

Yoshimura, Hirokazu, see also Howard, Robert.

Zinn, Robert, and Conrad C. Dahn, Variable
19 in NGC 5466: an anomalous Cepheid in
a globular cluster, Astron. J., 81, 527-533,
1976.

Zinn, Robert, and Leonard Searle, The masses
of the anomalous Cepheids in the Draco
system, Astrophys. J., 209, 734-747, 1976.

Zirin, Harold, Further observations of the
\10830 He line in stars and their significance
as a measure of stellar activity, Astrophys. J.,
208, 414-425, 1976.

Zirin, Harold, Production of a short-lived fila-
ment by a surge, Sol. Phys., 60, 399-404, 1976.

Zirin, H., and K. Tanaka, Prospectus for the
solar maximum, Sol. Phys., 47, 385-387, 1976.

Zirin, H., R. L. Moore, and J. Walter, Editors
of Procee flings of the Workshop: The Solar
Constant and the Earth's Atmosphere, Big
Bear Solar Observatory, May 1975, Sol. Phys.,
46, 377-409, 1976.

HALE OBSERVATORIES

191

STAFF AND ORGANIZATION

Dr. Eric E. Becklin resigned, effective
June 30, 1977, to accept a position as
scientist in charge of the new 3-meter
NASA infrared telescope at the Univer-
sity of Hawaii.

Dr. Robert J. Brucato was appointed
Scientific Officer for the Las Campanas
Observatory.

Dr. James E. Gunn and Dr. George W.
Preston were elected to membership in
the National Academy of Sciences.

Research Division

Staff Members

Halton C. Arp

Horace W. Babcock, Director

Eric E. Becklini

Jesse L. Greenstein^

James E, Gunn^

Robert F. Howard

Jerome Kristian

Robert B. Leigh ton^

Guido Miinch^

Gerry Neugebauer^

J. Beverley Oke, Associate Director

S. Eric Persson

George W. Preston,

Assistant Director, Mount Wilson
Allan R. Sandage
Wallace L. W. Sargent^
Maarten Schmidt^
Leonard Searle
Stephen A. Shectman
Arthur H. Vaughan

James A. WestphaF
Harold Zirin^

Members Engaged in
Postretirement Studies

Alexander Pogo
Henrietta H. Swope
Ohn C. Wilson

Staff Associates

Robert P. Brucato, Scientific Officer, Las

Campanas Observatory
Michael W. Werner®

Senior Research Fellows

Ronald Moore
Robert J. Zinn^^

Research Fellows

William M. Adams
A. Ger de Bruyn^i
Alan M. Dressler^^
Daniel Y. Gezari^^
Richard F. Green
Eduardo Hardyii-i^
Mark Hartoogii-i*
Gordon J. Hurford
Vincent Icke
Frank Israel
Stephen L. Knapp^^
Daniel Kunth
Kenneth A. Marsh^^
Larry D. Petro^^

1 Resigned June 30, 1977.

2 Lee A. DuBridge Professor of Astrophysics,
California Institute of Technology.

2 Professor of Astronomy, California Institute
of Technology.

* Professor of Astronomy ; Chairman of the
Division of Physics, Mathematics, and Astron-
omy, California Institute of Technology.

^ Professor of Physics, California Institute of
Technology.

6 Professor of Astronomy ; Executive Officer
for Astronomy, Cahfamia Institute of Tech-
nology.

"^ Associate Professor of Planetary Science,
California Institute of Technology.

8 Chief Astronomer of Big Bear Observatory ;
Professor of Astrophysics, California Institute
of Technology.

^ Assistant Professor of Physics, California
Institute of Technology.

10 Las Campanas Observatory Fellow.

11 Carnegie Fellow.

12 Fellowship ended November 30, 1976.

13 Fellowship ended December 31. 1976.
1^ Fellowship ended August 30, 1976.

15 Fellowship ended April 1, 1977.

16 Fellowship ended September 1, 1976.

192

CARNEGIE INSTITUTION

Douglai: 0. Richstone^^
Jack W. Sulentic
Trinh X. Tluian^®
Altluw Wilkinson
Peter Wilkinson

Carnegie-Chilean Fellows

Guido Garay
Maria Teresa Riiiz^®

Librarians

Helen Z. Knudsen
Nan W. Schow

Senior Research Assistants

Grace V. Knox
Charles T. Kowal
A. Louise Lowen^^

Research Assistants

John M. Adkins
John E. Boyden
Ken D. ClaVdy
Basil N. Katem
Margaret Katz
Stephen P. Padilla
Frances Y. C. Tang

Student Observers

Steven V. W. Beckwith
Kirk Borne
Todd Boroson^^
France Cordova
David J. Diner
Jonathan H. Elias
John G. Hoessel
John P. Huchra2i
Steven Kent
Barn- J. LaBonte
Jorge Melnick^^

Daniel Xadeau
William C. Priedhorsky

^' Fellowship ended Decomher 1, 1976.
i** Fellowship ended July 1, 1976.
i» Fellowship ended August 15, 1976.

20 Retired June 30, 1977.

21 Graduated June 11, 1976.

22 Graduated June 10, 1977.

Douglas M. Rabin
Russell 0. Redman
Anneila I. Sargent-
Donald P. Schneider
William L. Sebok
David Sholle22
Richard J. Terrile
Richard Wade
Peter J. Young

Photographic Department
John R. Bedke, Photographer

Instrument Design and Construction

David A. Bell, Electronics Engineering Aide
Louis E. Beidler, Engineer^^
Lawrence E. Blakee, Supervisor,

Electronics Services
Barbara L. Dailey, Draftswoman
Lee B. Dennison, Design Draftsman^*
Stephen Doro, Machinist
Earle B. Emery, Research Engineer
Eugene B. Fair, Head Optician
Hannah Fox, Computing Analyst
Jerry T. Fridenberg,

Head, Astroelectronics Laboratory
C. L. Friswold, Head, Design Group
Robert D. Georgen, Machinist
Richard M. Goeden, Senior Engineer
John G. Golson, Optician
Simon Groesz, Electronics Specialist
Fred H. Harris,

Junior Electronics Technician
Melvin W. Johnson, Optician^^
Leroy M. Kimoto,

Senior Electronics Specialist
Erich R. Koch, Senior Electronics Engineer
Wilfred H. Leckie, Senior Draftsman
Wilham H. McLellan, Senior Engineer
Roger L. Minnix, Engineer
Frederick G. O'Neil, Shop Foreman
Richard A. Prout, Senior Engineer^^
Juan M. Sanchez,

Photographic Laboratory Technician
Orval A. Smith, Electronics Specialist
Edward H. Snoddy, Designer
Sharon Soltesz, Administrative Secretary
Douglas Sprague,

Senior Electronics Engineer

23 Terminated April 15, 1977.

24 Terminated December 3, 1976.

25 Retired June 30, 1977.

26 Resigned March 11, 1977.

HALE OBSERVATORIES

193

Merle R. Sweet, Supervisor,
Electronics Construction
David F. Thompson, Technical Assistant
Glenn A. Toennes, Electronics Technician
Timothy J. Tyler, Electrical Technicians"^
Virgil Z. Vaughan, Supervisor, Stockroom^^
Fehce Woodworth, Illustrator

Operation and Maintenance
Mount Wilson Observatory and Offices

Robert E. Cadman,

Mountain Superintendent
Herman E. Carpentier, Carpenter
Linda Chaffee, Clerk-Typist
Hugh T. Couch,

Superintendent, Buildings and Grounds
Helen S. Czaphcki, Typist-Editor
Bette DeSmet, Secretary
James Frazer, Night Assistant
Hazel M. Fulton, Head Stewardess
Joan Gantz, Assistant Librarian
Thomas S. Gregory, Solar Observer
Robert J. Grosdidier,

Electronics Technician
Eugene L. Hancock, Senior Night Assistant
Mary Hark, Stewardess
Jeanne M. Knight, Bookkeeper
Howard H. Lanning, Night Assistant
Jose Lopez-Tiana, Purchasing Clerk
Ernest 0. Lorenz,

Assistant Mountain Superintendent
Ethel Marszalek, Custodian^^
Frank Perez, Mechanic
Dolores Sahlin, Typist-Telephone Operator
Glen Sanger, Senior Custodian
William D. St. John, Chauffeur
Frank Trylko, Custodian
Frederick P. Woodson, Business Manager
Gary Yanik, Electronics Engineer

Palomar Observatory and
Robinson Laboratory

Ranney G. Adams, Night Assistant
Albert R. Andrews, Maintenance Mechanic
Bradley N. Bailey,

Night Assistant and Junior Technician
Ray L. Ballard,

Senior Administrative Assistant
Stephen A. Barry, Maintenance Mechanic^^

Donald C. Bates,

Assistant Mountain Superintendent-^^
Jan Adriaan Bruin.^m,'), Painter
Maria J. Bruinsma, Hous<^'keeping Aide
Juan R. Carrasco,

Night Assistant and Mechanic
Lily D. Carrasco, Housekeeping Aide
Rita A. Ewing, Secretary
Liselotte M. Hauck,

Senior Administrative Secretary
Helen Holloway, Administrative Secretary
Dorothy J. Howard,

Administrative Secretary
Willie D. Jones, Custodian
Kevin M. Jordan, Maintenance Mechanic
Joyce E. Keeble, Secretary
Taras Kiceniuk, Mountain Superintendent
J. Luz Lara, Maintenance Mechanic
Marilynne J. Rice,

Senior Administrative Secretary
Elsa-Brita Titchenell,

Administrative Secretary
Gary M. Tuton, Senior Night Assistant
Paul Van Ligten, Electrician
Ruth E. Weaver, Administrative Secretari^
Larry L. Wickern, Assistant Superintendent
Larry K. Williams, Maintenance Mechanic
Dorothy Williams, Cook
Mary R. Yates, Housekeeping Aide^^
Barbara A. Zimmerman,

Computing Analyst-

Big Bear Solar Observatory

Alberta R. Altman, Secretary
Jack R. Klemroth,

Solar Observing Assistant
Eugene H. Longbrake, Superintendent
Charles F. Mason, General Machinist^^
Walter M. Nagao, Custodian
Jeff B. Nenow, Solar Observing Assistant
Alan P. Patterson, Associate Solar Scientist
Ow^en Phairis, Solar Observing Assistant

Las Caynpanas Observatory

Maynard Clark,

Senior Electronics Technician
S. Thomas Couch,

Administrative Assist ant ^^
Ljubomir Papic, Mountain Superintendent
Manfred Wagner, Administrative Manager

27 Terminated September 30, 1976.

28 Retired November 22, 1976.

29 Terminated June 30, 1977.

30 Resigned December 9, 1976.

31 Resigned October 21, 1976.

32 Resigned May 23, 1977.

33 Retired December 31, 1976.

34 Terminated April 15, 1977.

194

CARNEGIE INSTITUTION

H ^

1—1 0)

(N '^

o
f— I w

o d

O o
cS CO
■^^ J3

>. o

O (1)
o M

« _5

TO 75

■ d

^ g
-^ 5

2 d

*^ bC

o) d
^S

a ■«

>^ o

X 0)

■^ _d

■*'^

^ o .2

^ -^ bC

2 o d

tH 03 o

'd '^ X

d 0) X

O -M -^^

CC rd M

_d .22 9
H cu O

- d fc

"^ "^ -d

HALE OBSERVATORIES

195

Oi

<

a
a

O
3

a
c
O

196

CARNEGIE INSTITUTION

<N

<

o

o

03

a

0)

a

CO

o

d

CO

O

o

Hi

d

a

o

&

d
o

02
0)

CO

Si

HALE OBSERVATORIES

197

o

'5:

o

o

O

o

O

G

o

198

CARNEGIE INSTITUTION

73

O

d

T3

bD

05

C<J

p.

<1

d
o

CD

o

4^

d
o

d

bD
O

O
ft

s

CO

o

o

:zi

>.

X

^

Is

bD

t(-i

u

d

^

-d

9

22

_bC

a;

X3

H

Department of Plant Biology

Stanford, California

Winslow R. Briggs

Director

Carnegie Institution of Washington Year Book 76, 1976-1977

Contents

Introduction (Briggs) 203

Spectral studies of P700-chlorophyll a-protein complexes (Brown) 209

Action Spectra (French) 212

Photosynthetic electron flow from photosystem II requires membrane-bound bicar-
bonate (Stemler) 217

The effect of temperature on the physical phase of chloroplast membrane lipids and
photosynthesis (Fork) 220

Introduction 220

Lipid phase changes in lettuce and spinach chloroplasts at sub-zero temperatures

(Murata and Fork) 220

Studies on the effect of transition of the physical phase of membrane lipids on elec-
tron transport in the extreme thermophile Synechococcus lividus (Fork and
Murata) 222

The Hght-induced carotenoid shift in Cyanidium and in higher plant leaves as an

indicator of phase changes in chloroplast membrane lipids (Murata and Fork) 226

The rate of proton gradient decay in chloroplasts as an indicator of membrane

parameters (Avron and Fork) 231

Proton gradients as possible intermediary energy transducers during ATP-driven

reverse electron flow in chloroplasts (Avron and Schreiber) 236

ATP-induced chlorophyll luminescence in isolated spinach chloroplasts (Schreiber and

Avron) 236

Studies on DNA sequence organization (Murray, Preisler, and Thompson) 240

Single-copy DNA sequence comparisons in Atriplex (Belford and Thompson) 246

Interspecific hybridization of fern DNA (Stein and Thompson) 252

Contaminants affecting plant DNA reassociation (Murray and Thompson) 255

Single-strand DNA fragment length determination by alkaline agarose electrophoresis:

apphcations for DNA renaturation studies (Murray and Thompson) 259

In vitro labeling of single-stranded DNA (Murray, Belford, and Thompson) 262

Restriction endonuclease analysis of Agrobacterium plasmids (Rogler, Thompson,

and Cohen) 267

The use of two-wavelength microspectrophotometry to quantify light-induced changes
in the intracellular distribution of phytochrome determined b}^ immunoc}i;o-
chemical locahzation (Britz, Mackenzie, and Briggs) 274

A recording microphotometer for measurement of chloroplast orientation movements

in single algal filaments (Blatt and Briggs) 278

The auxin hypothesis of light-inhibited cell elongation (Vanderhoef and Briggs) .... 282

Kinetic and spectral studies on red-hght-induced suppression of mesocotyl elongation

in maize (Vanderhoef and Briggs) 283

Red- and Blue-Light-Induced Irreversible Absorbance Changes in Particulate Frac-
tions of Com Coleoptiles (Britz, Widell, and Briggs) 286

Methylene blue-mediated red-light photorediiction of cytochromes in particulate

fractions of corn and Xcurospora (Britz, Schrott, Widell, Brain, and Briggs) .... 289

Blue-light-induced absorbance changes in membrane fractions from Pisuin sativum

(Galston, Britz, and Briggs) 293

Correlative studies of hght sensitivity and cytochrome content in Neurospora crassa

(Brain. Woodward, and Briggs) 295

Solubilized membrane-bound auxin receptors from corn (Cross, Dohrmann, Briggs,
and Ray) 299

External labeling of membranes in intact plant tissues (Cross and Briggs) 303

Ivinetic and ste<ady-state studies of ^^''O fixation into photorespiratory intermediates
by intact leaves of C3 species (Berry, Osmond, and Lorimer) 307

Incorporation of [^^0] oxygen into glycolate by intact isolated chloroplasts (Lorimer,
Krause, and Berry) 314

Induction of COo-independent photosynthetic O2 evolution in Dunaliella salina

(Kaplan, Schreiber, and Avron) 316

A proton gradient in intact cells of Dunaliella salina (Kaplan and Schreiber) 320

Heat-induced cholorophyll fluorescence changes in intact leaves correlated with dam-
age of the photosynthetic apparatus (Schreiber and Berry) 323

Photosynthetic acclimation to temperature and water stress in the desert shrub Larrea

divaricata (Mooney, Bjorkman, and CoUatz) 328

Photosynthetic acclimation to temperature in Larrea divaricata: Light harvesting
efficiency and capacity of photosynthetic electron transport reactions (Armond,
Schreiber, and Bjorkman) 335

Heat-induced changes in chlorophyll fluorescence and related heat damage at the

pigment level (Schreiber and Armond) 341

Thermal stability of photosjnithetic enzymes in heat- and cool-adapted C4 species

(Bjorkman and Badger) 346

Studies on the kinetic mechanism of ribulose-l,5-bisphosphate carboxylase and
oxygenase reactions, with particular reference to the effect of temperature on
kinetic parameters (Badger and Collatz) 355

The internal CO2 pool of Chlamydomonas reinhardtii: Response to external CO2

(Badger, Kaplan, and Berry) 362

Adaptive value of leaf hairs in Encelia farinosa (Ehleringer) 367

Bibliography 369

Speeches 370

Personnel 374

INTRODUCTION

In January 1977, a visiting committee partment to hear Carnegie scientists
assembled by the Board of Trustees spent exj)lain their work and to see them
the better part of two days at the De- demonstrate some of their equipment,
partment of Plant Biology. Three of the Other classes, from Stanford and else-
trustees and three plant scientists not where, used the field stations in the
associated with Carnegie looked into Sierra Nevada — Mather and Timberline
every facet of the Department — its teach- — for systematic or ecological studies,
ing, its finances, and its science. The Department continued to make an

Except for a few gentle caveats, the important contribution to graduate edu-

committee's report was very favorable, cation through the participation of staff

It lauded the Department for its unique members in thesis committees and the

combination of informality, opportunity appointment of graduate students as

for interaction, and adequate support fellows or assistants. Within the De-

without pressure for short-term publica- partment, the mix of graduate students

tion. This is the Carnegie approach to with postdoctoral fellows and more senior

basic research, and it has consistently visitors is nearly optimal for effective

led to outstanding results in plant science, interaction.

Later in the year we were also greatly In summarizing some of the research
encouraged and gratified by a grant of accomplishments of the past year, I feel
$600,000 from the Andrew Mellon Foun- the usual uneasiness at attempting to
dation for research on photosynthesis, single out highlights from a mass of
The Mellon grant will enable us to in- sound research. The first section deals
tensify our efforts to understand this primarily with basic aspects of photo-
complex and vital process. synthesis at the mechanistic level, the

This year, once again, every lab desk second with nucleic acids, the third with

was occupied. Visitors ranging from un- the role of light and hormones in plant

dergraduates to senior scientists on sab- growth, and the final section returns to

batical exchanged information and ideas photosynthesis, this time primarily in the

about a variety of research problems, context of adaptation to stress.

Their collaboration resulted in a number Several of the articles report either the

of articles, which appear later in this development of techniques or the design

Report. of special equipment. For example, C.

The Department continues to play an Stacy French, continuing his long-stand-
important teaching role. Faculty mem- ing interest in action spectroscopy, has
bers were repeatedly called on to give found an ingenious way of measuring
guest lectures at Stanford University, steady-state dye concentration in light
and visiting fellows occasionally par- and dark in a system containing dye and
ticipated as well. Most of the younger photosynthetically active particles. The
scientists had an opportunity to present beauty of the method is that it averages
their work for critical review at seminars out ''noise" and permits far greater ac-
with Carnegie and Stanford botanists, curacy than was possible before. ]\Iichael
Carnegie staff and fellows also partici- Blatt, a graduate student working with
pated with Stanford faculty in an ad- Briggs, has developed a technique for
vanced seminar course on plant move- studying the influence of light on chlo-
ments. Three groups of students from roplast movement in single algal fila-
San Jose State University and San Fran- ments. The method utilizes two optic
Cisco State University came to the De- fibers: One carries both actinic and

203

204 CARNEGIE INSTITUTION

measuring beams to the filament ; and photosynthesis group, continued his

the other, equipped with a filter to screen studies on the role of the bicarbonate ion

out the actinic light, carries light trans- in the oxygen-evolving portion of system

mitted through the filament to a photo- II in photosynthesis. He showed that

diode for continuous monitoring. there was a very close correspondence

Britz and Mackenzie, also with Briggs, between the binding of bicarbonate to
adapted a two-wavelength microspectro- thylakoids and oxygen-evolving capacity,
photometric method to quantify the that the bicarbonate, once bound, was
fractional area covered by pigment in a not freely exchangeable, and that the
system in which the distribution is non- binding site was perhaps not far from
homogeneous. They applied the technique the site of action of the system II in-
to determine indirectly the extent of hibitor DCMU.

light -induced redistribution of the pho- Fork has continued his studies of the

tomorphogenic pigment phytochrome. influence on photosynthetic activities of

Murray and Thompson discovered a physical phase transitions of thylakoid
contaminant that can greatly accelerate membrane lipids. In chloroplasts of let-
plant DXA reassociation, and developed tuce and spinach that have high concen-
techniques to eliminate it, thereby re- trations of unsaturated fatty acids, these
moving a serious and puzzling source of phase transitions occurred well below
variability in reassociation experiments. 0°C. He and Murata obtained evidence
They also modified an electrophoresis that increasing either magnesium or
technique to give extremely accurate potassium concentration in chloroplasts
measurement of the lengths of single- from spinach or lettuce markedly de-
stranded DNA fragments. Finally, Mur- pressed phase transition temperatures. In
ray and Thompson modified a technique contrast to the lettuce and spinach, in
for labeling single-stranded DNA so that which the phase transition temperatures
far longer and. more complex DNA frag- were all below 0°C, the thermophilic
ments could be labeled than previously, blue-green alga Synechococcus lividus,
All three of these studies of plant nucleic which is capable of growing photosyn-
acids sharpen the tools for investigation thetically at 75°C, showed a phase transi-
of the properties of the plant genome in tion near 43°C when grown at tempera-
organization, evolution, and expression tures above 55°C. When the algae were
during development. grown at lower temperatures, lower phase

Brown has made progress in char- transitions were seen. Consistent with the
acterizing the reaction center chlorophyll results obtained earlier with the blue-
complex from higher plants and algae. In green alga Anacystis, the photosynthetic
addition to showing electron transfer to electron transport capacity of Synecho-
cytochrome 6(5 under the appropriate con- coccus declined sharply below the phase
ditions of pH and illumination, she ob- transition temperature,
tained evidence from various sources for Murata and Fork also found that
a tightly bound carotenoid in the com- phase transitions could be detected by
plexes and determined the relationships measuring the dark decay of the light-
between differently fluorescing systems, induced carotenoid change. These studies
The interfacing of the Hewlett-Packard showed that chilling-sensitive plants had
21 16 computer with the spectrofluorimeter a phase transition above 0°C that varied
greatly increased the resolution of. the with growth temperature, while chilling-
fluorescence studies, particularly in re- resistant plants did not. Finally, Fork
solving fluorescence bands from back- and Avron studied the influence of tem-
grounfl noise when total fluorescence was perature on the rate of proton efflux
extremely low. through thylakoid membranes by mon-

Stemler, a postdoctoral fellow with the itoring the fluorescence of 9-aminoacri-

DEPARTMENT OF PLANT BIOLOGY 205

dine. They found the rate to be sharply will help elucidate the functional sig-
temperature-dependent and an excellent nificance of the organization,
indicator of phase transitions and other Bel ford has greatly refined and ex-
temperature-dependent membrane pa- tended her techniques for investigating
rameters. Their technique also showed relationships among species of the genus
the clear differences between chilling- Atriplex by a study of single-copy DNA
sensitive and chilling-insensitive higher homology. The greater precision and ad-
plant chloroplasts. ditional species examined strengthen the

Concluding the reports of studies on earlier conclusion that present-day species

photosynthesis per se are two articles utilizing the C4 photosynthesis pathway

by Avron (on sabbatical leave from the did not arise independently, but prob-

Weizmann Institute) and Schreiber. ably diverged from a common C4 stock

These describe how 9-aminoacridine is early in the history of the genus,

used to monitor proton pumping and Stein extended an evolutionary study

efflux to determine how closely proton of the fern genus Osmunda to resolve an

pumping parallels changes in the reduc- inconsistency between several recent

tion of the primary system II electron DNA hybridization experiments and

acceptor when reverse electron transport some earlier results. The sensitivity and

is driven by exogenous ATP. Under a reproducibility of the inter-species com-

wide variety of conditions, the buildup parisons by DNA hybridization allowed

of protons in the thylakoids remarkably the conclusion that one collection made

paralleled the reduction of the primary in 1973 was either from a natural hybrid

acceptor, thus strengthening the chemi- population or contained tissue from a

osmotic hypothesis for energy transduc- related, morphologically similar species,

tion in chloroplasts. Avron and Schreiber The results also indicated that the tech-

also found conditions under which elec- nique permits discrimination between

tron transport could be driven backward closely related species, and can possibly

with ATP to the photosystem II reaction even discern hybrids. This observation

centers, and they were able to monitor suggests that detailed studies of repeated

the process by chlorophyll luminescence, sequence evolution in Atriplex species

In addition to the technical studies may soon be possible,

mentioned earlier, Thompson's group has Rogler, with Thompson and Cohen,

made substantial progress in several basic completed a careful comparative study

areas. Murray and Preisler, together with of restriction enzyme fragments from

Thompson, have obtained some solid purified plasmids from both tumor-induc-

comparative data on sequence organiza- ing and non-tumor-inducing strains of

tion in the pea and mung bean genomes, the crown gall bacterium Agrobacterium

Both plants, in common with most ani- tumefaciens. It is clear that there are

mals, contain single-copy sequences in- more DNA sequences in common among

terspersed with repetitive sequences, the tumor-inducing plasmids than be-

Mung bean, however, has more single- tween them and the non-tumor-inducing

copy sequences than the pea has; and in ones, at least on the basis of DNA frag-

the mung bean the long-period intersper- ment size. It is also evident that classes

sion pattern is more common. Peas, hav- of a certain size are duplicated, though

ing more total repetitive DNA, exhibit whether or not the duplicates represent

much more short-period interspersion — identical sequences has not yet been re-

and may actually have no significant solved. As in other studies by the plant

long-period component at all. There are nucleic acid group, the resolving power

other differences, and it is clear that con- of the techniques has been greatly im-

tinued comparative study of both DNA proved in the past year,

sequence organization and transcription Vanderhoef (on sabbatical leave from

206 CARNEGIE INSTITUTION

the University of Illinois) and Briggs tions from another higher plant, the pea
completed a detailed study on the influ- seedling. Like corn and Neurospora, peas
ence of red light on elongation of the yielded a membrane fraction enriched in
first internode of germinating corn seed- the photoreducible system. It also could
lings. They found that light treatment be reduced by red light if methylene blue
brings about a dramatic inhibition of was present. The pea system permitted
growth, and that the effect can be com- measurement of reversible light-induced
pletely reversed by adding the plant hor- flavin reduction alone, in the absence of
mone auxin. The inhibition shows two any interaction with the cytochrome, and
phases, one in response to extremely low yielded a difference spectrum consistent
doses of light and the other requiring with that for a flavoprotein.
much larger doses. An action spectrum Brain, Woodward, and Briggs obtained
for the first phase shows a peak at about evidence that Neurospora mutants de-
665 nm. indicative of the well-known ficient in 6-type cytochromes are also
photomorphogenic pigment phytochrome, deficient in light-reducible cytochrome in
but there is a second sharp peak at 640 membrane fractions and have signifi-
nm which cannot be attributed to phyto- cantly impaired photosensitivity. That
chrome. Earlier workers had noted the the reduced photosensitivity was not
peak but otherwise ignored it. Britz, simply an indirect result of lowered
Widell. and Briggs have detected in respiratory capacity was suggested by
membrane fractions from corn a pigment the appearance of normal photoreducible
absorbing near 630 nm that is effectively cytochrome responses in membrane prep-
bleached by both red and blue light, arations from another kind of respiratory
They present evidence that this pigment mutant with normal cytochromes. These
might be the 640-nm photoreceptor (pig- results, together with the pea and methy-
ments frequently show spectral shifts lene blue studies, are the strongest to
upon extraction). Whatever the outcome date implicating the fiavoprotein-cyto-
of further studies, there is now reason to chrome complex in the blue-light photo-
believe that phytochrome may not be the reception process.

only photomorphogenically active pho- Cross, continuing with Dohrmann and

toreceptor absorbing in the red region of Ray a project begun by Dohrmann last

the spectrum. year, succeeded in the solubilization and

Since methylene blue was known to partial purification of a macromolecule

endow normally blue- and ultraviolet- that binds the plant hormone analog

sensitive processes in several fungi with naphthylene acetic acid. He was able to

red-light sensitivity, Britz, with Schrott, separate it clearly from ATPase activity

Widell, and Brain, investigated whether by column chromatography, thereby

the dye would also confer red-light sensi- weakening the hypothesis that auxin acts

tivity on the normally flavin-mediated by binding and somehow affecting the

photoreduction of a 6-type cytochrome, activity of a membrane-bound ATPase.

thought to be a candidate for the blue- Cross also demonstrated (contrary to the

ultraviolet photoreceptor in higher plants literature and our own expectations) that

and fungi. Methylene blue did indeed the fluorogenic dye fluorescamine is not

mediate light-induced transfer of elec- a specific label for cell surfaces but that

trons to the appropriate cytochrome, and it shows a high affinity for any cellular

showed a fairly high degree of specificity membrane and presumably reacts with

in com and Neurospora membrane frac- the free amino groups of such molecules

tions. as phosphatidyl ethanolamine. Definitive

Galston, visiting briefly from Yale, identification of a specific membrane

searched with Britz for light-inducible fraction such as the plasma membrane

cytochrome reduction in membrane frac- has not been achieved, but current studies

DEPARTMENT OF PLANT BIOLOGY 207

with fluorescent mercury compounds fluorescence. They were able not only

seem promising. to rank the plants in terms of heat sta-

Berry, who has been on sabbatical bility but also to distinguish differences

leave from the Department for the past in heat tolerance in specimens of a single

year, has been investigating the origin species grown at different temperatures.

of glycolate and other photorespiratory They also showed that by a mechanism

intermediates in collaboration with Os- currently not understood, high light in-

mond, Lorimer, and Krause. They con- tensity can provide substantial protection

elude that it is extremely unlikely that against heat damage,

glycolate arises in significant amounts A series of studies on the desert shrub

from a pathway other than the oxygena- Larrea divaricata illustrates, among other

tion of ribulose-l,5-bisphosphate car- things, the degree of interaction among

boxylase. Understanding the origin of research groups in the Department and

glycolate should help in the understand- the integrated approach that such inter-

ing of environmental control of photo- action can produce. Mooney, Bjorkman,

respiration. and Collatz studied photosynthesis in

In a study of the highly salt-tolerant Larrea seedlings from Death Valley
alga Dunaliella, Kaplan, with Avron grown under three temperature regimes.
and Schreiber, studied some very inter- It was clear that the temperature for
esting responses to increases in sodium optimum photosynthesis was higher when
concentration. They showed that addition the plants were grown at higher tempera-
of salt brought about a sharp increase tures, and that the disparity was unre-
in oxygen evolution, even in the presence lated to differences either in the CO2-
of extremely low CO2. It was already fixing enzyme or in some aspect of pho-
known that such treatment would greatly torespiration. They then examined the
increase the glycerol concentration within effect of water stress on the quantum
the cell. The results were consistent with yield of photosynthesis for Larrea plants
the hypothesis that the increase in sodium grown under ample water supply, com-
concentration resulted in a diversion of pared to the effect on plants from Death
the NADH and ATP from photosynthesis Valley. When the stress was — 36 bar,
to the production of the glycerol from the quantum efficiency in the first group
stored carbohydrate reserves. Kaplan dropped by half. What was startling,
and Schreiber then used the same dye however, was that the same species,
mentioned earlier, 9-aminoacridine, to growing in Death Valley, showed a nor-
show the existence of a light-driven mal quantum efficiency and little change
sodium-dependent proton uptake system, between — 20 and almost — 50 bar. The
probably involving an ATPase. The re- results show not only that Larrea can
suits could explain why Dunaliella can photosynthesize effectively under re-
tolerate such dramatic changes in sodium markable drought stress in nature, but
concentration without taking up any also that one must be very careful not to
sodium ions. extrapolate from greenhouse and growth-

The physiological ecology group has chamber experiments without doing care-
done extensive experimentation upon the ful field studies as well,
effect of temperature on whole plants or Armond, Schreiber, and Bjorkman next
plant parts. Schreiber and Berry, work- isolated chloroplasts from the Larrea
ing with leaves of plants with large dif- plants grown under the three tempera-
ferences in heat tolerance, monitored ture regimes and investigated fluorescence
changes in fluorescence both with gradu- properties of the chloroplasts with heat-
ally increasing temperature and also with ing. Plants grown at the highest tempera-
a sudden temperature jump under con- ture regime showed little change in fluo-
ditions yielding either fixed or variable rescence spectra with heating, though

20S CARNEGIE INSTITUTION

they had the hirgest photosynthetic between species adapted to high tempera-
units. Chlorophists from ph\nts grown on tiires and those adapted to low, there
the other two regimes, however, showed are differences in enzyme stability as well,
marked heat damage to energy transfer Badger and Collatz have undertaken
from clilorophyll b to chlorophyll a at a detailed kinetic study of the enzyme
temperatures well below those required ribulose-l,5-bisphosphatase carboxylase
to inactivat<^ electron transport. Clearly, to learn something of the carboxylase
pigment organization is a weak point and oxygenase reaction mechanisms. Al-
in temperature stability. Furthermore, though the results are incomplete, they
Larrea plants grown at a higher tempera- suggest that neither reaction is a strictly
ture can adjust to improve the stability ordered one, with the COo or O2 always
of the pigment organization. binding to the enzyme after the ribulose-

More detailed experiments by Schreiber 1,5-bisphosphate. Badger, Kaplan, and
and Armond showed heat damage to Berry have also continued work begun
transfer of energy among forms of chlo- last year on the mechanism by which
rophyll a as well, and also showed that the alga Chlamydomonas reinhardtii
during heat inactivation those reaction adapts to growth on low concentrations
centers which were still active were fully of CO2. The results clearly show that
active and the photosynthetic unit size the adaptation involves development of a
imchanged. By helping to describe how system to concentrate CO2 from the ex-
one plant copes with drought stress, these ternal medium. This system is unrelated
studies have added substantially to our to C4 photosynthesis in higher plants,
knowledge of the nature of heat damage Ehleringer, as part of a wider study
in plants. of leaf hairs in the genus Encelia, has

In another study on heat damage, done a combination of experimentation
Bjorkman and Badger investigated the and modeling which show that the leaf
thermal stability of 14 photosynthetic hairs in this plant are highly adaptive,
enzymes from the high-temperature- Without reflective leaf hairs, the plant
adapted C4 species Tidestromia oblongi- would be unable to survive the extreme
folia and the cool-temperate C4 species temperatures of its habitat.
Atriplex sabulosa. Of the 14 enzymes Highlights of research for the year in-
studied, four had heat stabilities that elude the extensive temperature studies
were identical or similar in the two spe- in Fork's laboratory, the substantial tech-
cies, while another six showed difTerences nical improvements and increased resolu-
of from 6 to 10°C. In more than half tion developed in Thompson's laboratory,
the enzymes investigated, heat inactiva- the increasing evidence from Briggs'
tion of the Atriplex enzyme occurred at laboratory that a flavoprotein-cyto-
temperatures lower than those affecting chrome complex is directly involved in
the overall rate of photosynthesis in some way in photoreception of blue and
Tidestromia but usually higher than ultraviolet light, and the extensive studies
those required to inhibit photosynthesis of adaptation to temperature stress by
in intact Atriplex. However, in both Bjorkman's group.

plants there were certain enzymes that It should also be mentioned that Nobs

showed heat inactivation near the tem- and Hiesey are assembling for publica-

perature at which overall photosynthesis tion the results of an enormous program

is inactivated. Many of these are enzymes of hybridization and transplantation of

that require light activation. Thus, in grasses carried out years ago. This work

addition to the known differences in heat is a welcome addition to the five volumes

stability of system II electron transport of Experimental Studies on the Nature

and of noncyclic photophosphorylation 0/ Species.

DEPARTMENT OF PLANT BIOLOGY

209

SPECTRAL STUDIES OF P700-CHLOROPHYLL
a-PROTEIN COMPLEXES

Jeanette S. Brown

The basic mechanism by which energy
is converted in all plants from sunlight
absorbed by chlorophyll to chemical po-
tential is one of our most important
scientific problems. One approach to
this problem is to disrupt chloroplasts
(the chlorophyll-containing organelles in
which photosynthesis occurs) and to at-
tempt to extract the smallest photo-
chemically active piece. The most prom-
ising of the ^'pieces" thus far obtained is
the chlorophyll-protein complex, CPI,
which contains the reaction center of
photosystem I, P700. P700 is a special
form of chlorophyll u, probably a dimer,
which can be reversibly oxidized by visi-
ble light. CPI obtained by detergent
solubilization of chloroplast membranes
and separation by hydroxylapatite chro-
matography contains between 20 and 40
antenna chlorophyll a molecules in addi-
tion to the P700, depending upon details
of the extraction procedure. Two cyto-
chromes and an iron-sulfur protein nor-
mally co-chromatograph with the chlo-
rophyll-protein and a small amount of
carotenoid pigment. The chlorophyll-
protein structure of the basic complex is
not known, but a model proposed by
Thornber et at. (1977) suggests that the
P700 may be attached to one polypeptide
of a trimer having a total molecular
weight of 160 kilodaltons.

We have continued our spectroscopic
studies of this P700-chlorophyll-protein
complex isolated for the most part from
spinach chloroplasts and a yellow-green
and a blue-green alga. Fluorescence-
excitation spectra have given information
about energy transfer within the pigment
complex, and the dark reduction kinetics
of P700 by ascorbate have been clarified.

Fluorescence Excitation and
Emission of CPI*

An emission spectrum of a CPI prepa-
ration from spinach having a chlorophyll
to P700 ratio of 32 was shown in Year
Book 75, Fig. 47. An explanation for the
unusual emission maximum of this spec-
trum at 696 nm was suggested: namely
that the oxidized dimer of P700 fChl+ •
Chi) may absorb at about 690 nm and
fluoresce near 696 nm. Philipson et al.
(1972) suggested a similar hypothesis to
explain their circular dichroism results,
which indicated that exciton interaction
between chlorophylls of a P700 dimer
may cause absorption bands at 683 and
697 nm. Upon oxidation these bands
would disappear and a new band would
appear near 686 nm of the unoxidized
chlorophyll. This 686-nm absorption band
could be the source of the 696-nm emis-
sion maximum at low temperature.

The emission spectra reported last year
were measured with exciting light cen-
tered at 438 nm and a slit-wndth of 10
nm. A comparison of excitation spectra
for emission from spinach CPI at 696 or
670 nm (Fig. 1) shows that absorption
near 450 nm preferentially excites the
696 nm emission. The main excitation
bands near 440 and 420 nm correspond
to maxima of the absorption spectrum,
but no band can be distinguished on the
steep slope of the absorption spectrum
near 450 nm. When Triton is added to a
CPI preparation, emission at 650 and 670
nm is increased from solubilized chlo-
rophylls 6 and a, respectively. The emis-
sion near 650 nm is excited bv 465-nm

* CPI = P700-chlorophyll a-protein.

210

CARNEGIE INSTITUTION

T 1 1 1 1 r

360 380 400 420 440 460 480

640 660 680 700 720 740

Wavelength, nm

Fig. 1. Fluorescence excitation and emission (— 196°C) and absorption spectra (— 180°C) of
spinach CPI, chlorophyll/P700 = 30. For excitation spectra both sHt widths = 6 nm. For
emission spectra excitation slit widths = 10 nm, emission slit widths =r 3 nm.

light and the 670-nm emission by light
absorbed by chlorophyll a below 440 nm.
Even though only traces of these solu-
bilized chlorophylls are formed, they have
a very high fluorescence yield compared
to the pigment bound to the protein
complexes.

In an attempt to understand the source
of the excitation band near 450 nm, the
low-temperature fluorescence of a photo-
system I centrifugal fraction of spinach
chloroplasts broken in a French press
was measured for compari.son with CPI.
The emission maximum characteristic of
this spinach fraction is highest near 730
nm at — 196^' (Brown, 1971). An exci-

tation spectrum for emission at 730 nm
showed maxima at 440, 446, 450, 464,
470, and 474 nm. Apparently light ab-
sorbed by chlorophyll a at 440, chloro-
phyll b near 470, and carotenoid pigments
at the other wavelengths passes by reso-
nance energy transfer to the form of chlo-
rophyll a absorbing the longest wave-
lengths near 704 nm from which it is
emitted. A separate band near 696 nm
was not seen. Probably the Triton treat-
ment used to extract CPI from the mem-
brane solubilizes the long-wavelength,
aggregated chlorophyll forms (Ca692 and
Ca704) and thereby reduces the long-
wavelength fluorescence (>725nm) from

DEPARTMENT OF PLANT BIOLOGY

211

CPI. This effect may not occur during
chloroplast particle preparations.

A comparative study of chlorophyll-
protein complexes isolated from different
kinds of algae has continued with the
yellow-green Bumilleriopsis filiformis and
blue-green Anabaena * cylindrica. Pre-
liminary results with Bumilleriopsis indi-
cate that a chlorophyll complex analogous
in its chromatographic elution pattern to
spinach CPI can be obtained. The
chemically reduced minus oxidized ab-
sorption difference spectrum of this algal
complex is similar to that of P700-en-
riched higher plant fractions and showed
a chlorophyll-to-P700 ratio of about 45.
However, only a comparatively small, re-
versible light-induced absorption change
near 700 nm could be observed, and it
had very slow kinetics. It may be that
the detergent treatment removes the pri-
mary electron acceptor from P700 in this
alga. The 696-nm maximum was present
in the fluorescence emission spectrum but
was shorter in comparison to the 683-nm
band than in the spinach CPI spectra.
A comparison of excitation spectra for
emission at these two wavelengths showed
a maximum at 457 nm preferential for
emission at 696 nm.

CPI preparations from Anabaena were
enriched in both light-oxidizable and
chemically oxidizable P700 with a ratio
of about 55 chlorophyll a molecules to
one P700. These preparations also showed
a distinct fluorescence emission maximum
at 696 nm ( — 196°C) that was preferen-
tially excited by absorption at 425 and
450, with an additional strong band at
480 nm.

Collectively these results indicate that
the chlorophyll-protein complex isolated
from photosynthetic membranes by de-
tergent treatment and enriched in the
photosystem I reaction center chloro-
phyll, P700, contains a tightly bound
carotenoid that is able to transfer its
absorbed energy to chlorophyll that
fluoresces at 696 nm. Results by Ji et al.
(1968) showed that the absorption of
/^-carotene in vivo is modified from that

in solution, probably by its attachment
to the chlorophyll phytol and lamellar
I)rotein. It is reasonable to assume that
CPI prej)arations from different .s[)ecies
may contain a similarly bound carote-
noid with slightly different absorption
maxima. A Soret band of an oxidized
P700 dimer absorbing near 450 nm could
also be the source of 696-nm emission.
It will be of interest to measure action
spectra for P700 j)hotooxidation by these
complexes.

Excitation spectra can be measured
with an instrument such as the Perkin-
Elmer MPF-3L Fluorimeter. The xenon-
arc actinic light source contains a number
of intense emission lines between 450 and
500 nm which can distort the true excita-
tion spectrum for pigments such as ca-
rotenoids and chlorophyll b, which absorb
in this spectral region. Recently we have
been able to nullify this artifact by using
the Hewlett-Packard 2116C computer to
digitize and store collected spectra. Since
a concentrated solution of Rhodamine B
has an essentially constant quantum
yield over the wavelength region from
200 to 600 nm, its measured excitation
spectrum is proportional to the quantum
flux of the actinic light. Excitation spec-
tra between 400 and 500 nm with differ-
ent excitation slit-widths for the emission
of Rhodamine B at 550 nm have been
collected and stored in the computer. An
experimental spectrum is divided by the
Rhodamine curve measured at the same
slit- width to give the corrected excitation
spectrum free of artifacts caused by the
lamp emission or other characteristics of
the equipment.

Effect of pH on the Reduction
Kinetics of P700 by Ascorbate

Since CPI does not contain any natural
reductant for P700, ascorbate is usually
added to reduce P700 in darkness after
it is photooxidized. Shown below are the
reduction half-times of photooxidized
P700 as a function of pH and ascorbate
concentration.

212

CARNEGIE INSTITUTION

Ascorbi\ie

Conoeniratiou

mM

Reduction
Half-time (sec)
pH S pH 10

10
50

100

24
10

4

3
<1
<1

Because ascorbate donates protons as
well as electrons when it is oxidized, it
can he a more efficient reductant at higher
pH. but whether this property can ac-
count completely for these kinetic differ-
ences shown in the table is not clear and
should be further investigated.

At higher pH and ascorbate concen-
trations the extent of P700 oxidation at
a given light intensity may be less be-
cause the rate of reduction may exceed
the rate of oxidation. The extent of oxi-
dation as measured by the decrease in
absorption of P700 at 697 or 430 nm
appeared to decrease with time in con-
tinuous actinic light. This decrease was
found to be the result of the depletion of
oxygen, the final electron acceptor, in
the cuvette. Methyl-viologen can act as
an electron carrier between oxidized P700
and oxygen and thereby increase the rate
of this reaction.

Coupled P700 Photooxidation and
Cytochrome be Reduction

The procedure of Shiozawa et al.
(1974) for preparing CPI by Triton-
solubilization of chloroplast membranes
and hydroxylapatite chromatography
permits the co-chromatography of cyto-
chromes / and be along with the P700-
enriched chlorophyll-protein. As noted

previously {Year Book 74, pp. 779-783;
Year Book 75, pp. 460-465) no light-
induced oxidation of these cytochromes
could be observed even with added plas-
tocyanin. However, when the absorption
change of P700 was measured at 430 nm
with 100 mM ascorbate at pH 10, a large
positive change occurred after the initial
rapid decrease in absorption in continu-
ous light. That this change was caused
by cytochrome be reduction was con-
firmed by observing a maximum at 563
nm in the light-minus-dark difference
spectrum. The reduction kinetics of the
cytochrome were very slow (ti/2 = 2-3
min) , indicating that the cytochrome was
only loosely bound to the chlorophyll-
protein. Anaerobic conditions favored the
reduction, and there was some indication
in the difference spectrum near 552 nm
that cytochrome / was also reduced. The
cytochrome could be reoxidized by mix-
ing with air.

References

Philipson, K. D., V. L. Sato, and K.

Sauer, Biochemistry 11, 4591-4595,

1972.
Brown, J. S., Methods Enzymol. 23, 477-

487, 1971.
Shiozawa, J. A., R. S. Alberte, and J. P.

Thornber, Arch. Biochem. Biophys.

165, 388-397, 1974.
Ji, T. H., J. L. Hess, and A. A. Benson,

Biochim. Biophys. Acta 150, 676-685,

1968.
Thornber, J. P., R. S. Alberte, F. A.

Hunter, J. A. Shiozawa, and K.-S. Kan,

Brookhaven Sym. Biol 28, 132-148,

1977.

ACTION SPECTRA

C. Stacy French

The work reported here started with of the two photochemical systems of

the hope of finding good material and green plant photosynthesis. The purpose

methods for measuring as precisely as was to compare the sum of the two action

possible separate action spectra for each spectra with the absorption spectrum of

DEPARTMENT OF PLANT BIOLOGY

213

the material to see if the absorption by
chlorophyll can be accounted for by the
activity of the two recognized photo-
chemical steps in the normal process of
photosynthesis — in other words, to con-
struct a balance sheet for the distribution
of absorbed energy between the two pho-
tosystems and to check for small amounts
of inactive absorption by certain forms
of chlorophyll.

The Nostoc preparations of Arnon
were considered especially promising for
such experiments because these very
small particles give an essentially clear
solution with strong activity for both
photosystems. Furthermore, the action
spectra for various steps in the electron
transport system in Nostoc particles have
been studied in detail by Fork, Hiyama,
and Ford {Year Book 73, pp. 725-738).
The stability of the frozen preparations
make them particularly convenient. This
work has been made possible through
gifts of the frozen particles from Pro-
fessor Arnon and Dr. Hiyama and
through frequent consultation with Drs.
Brown, Fork, and Avron. Much time has
been spent on technical trivialities such
as calibrations, investigations of sources
of error, computer programming, pre-
liminary experimentation, and methods
of increasing the measurable signal with-
out undue distortion of the data. The
optimum format for presentation of
action spectra has been given some at-
tention. Although the basic objective of
the work as stated above has not yet
been reached, a very satisfactory pro-
cedure has been devised for measuring
the action spectrum of DCIP oxidation.
The beauty of this method is that we
can measure the stationary-state dye
concentration instead of having to evalu-
ate rates. Thus the noise level is aver-
aged over a long period. Attempts may
be made in the future to use similar
dynamic equilibrium methods to study
other steps in the electron-transport
chain.

Methods

Measuring System

The exposures are made in a 1 X 1 cm
cuvette with a plastic insert giving a
sharply defined top to the 1-cm liquid
level. The whole cell is illuminated by
the measuring beam from one side and
by the actinic beam at a right angle to
the measuring beam. Behind the cell an
aluminum reflector returns 72% of the
transmitted actinic beam. A reference
measuring beam goes through an ad-
jacent cell filled with water. Both of these
measuring beams fall on silicon cells
whose outputs are amplified and the
ratios of the outputs determined by a
divider. The transmission of the sample
is recorded on an effective recorder scale
of 1000 inches with a calibrated variable
offset. Part of the actinic beam is re-
flected by a glass plate to another silicon
cell to monitor the actinic intensity,
which shows on one pen of a two-channel
recorder. When rates are to be deter-
mined, the actinic monitor output goes
first to an integrator, then to the recorder,
to give a measure of the intensity X time
as shown in Fig. 2.

The absorption curves of the Nostoc
material and the dye are used with the
measured transmission at 593 nm to cal-
culate the average actinic intensity in
the cell for each wavelength. This aver-
age intensity, including the light passing
through and reflected back, is used in-
stead of the incident intensity to calculate
the action spectrum. This makes it pos-
sible to use a high enough absorption to
give an adequate response without dis-
torting the results. The average intensity
for each pass is calculated as:

-.{

0.434 \

j(l _e-^/o-434)

E

where E is the optical density of the
mixture for the actinic wavelength. The
spectral half-width of the actinic light
from the monochromator was 4 nm, small
enough not to distort the action spectrum

214

CARNEGIE INSTITUTION

Fig. 2. The transmission of Nostoc particles
suspended in DCIP solutions. For 593 nm where
the dye absorption is high, transmission in-
creases on exposure to light that reduces the
dye. The integral of intensity X time of the
actinic beam is shown by the lower trace. The
hght efifect is determined by extrapolation of
the transmission drift before and after the
exposure.

detectably. This was checked with the
slit-width correction program of Jones
adapted to the Hewlett-Packard 2116
computer by Glenn Ford.

DCIP Reduction

Before the DCIP oxidation was studied,
an action spectrum for reduction of the
same dye was measured. This work was
a repetition of the study of Fork, Hiyama,
and Ford (Year Book 73, pp. 725-738)
to check out the modifications in the
procedure. For DCIP reduction the fol-
lowing reaction mixture was used at
20X': XaHPO, pH 6.4, 2.5 X lO-^M;
DCIP 3.33 X 10--^ M] DPC 4 X 10"^
M] MgCU 10-2 M; NaHC03 2 X 10-'^
M; sucrose 0.5 M; green material about
2.2 fig ' chl/ml, absorbing about 30% at
680 nm (higher in some experiments).
Initially the dye concentration drifted
toward an increase in color (reduction of

partially oxidized dye), and then after
half an hour of intermittent illumination
the dye was slowly oxidized. For the few
minutes before, during, and after an ex-
posure of about 1 minute, these rates of
drift were nearly constant. The change
in transmission due to the light exposure
was determined by extrapolation of the
''before" and "after" drift lines to the
center of the exposure time as shown in
Fig. 2. This is a somewhat less than
satisfactory procedure; it is nevertheless
preferable to measuring slopes.

DCIP Oxidation

Particles of photosynthetic material
in a solution of DCIP reduced to its
colorless form by an excess of ascorbate
wdll oxidize some of the dye to its colored
form (Vernon and Zaugg, 1960). Equality
is reached between the rate of the photo-
oxidation and the rate of dye reduction
by the ascorbate in about a minute at
10°C, with a few percent of the dye in
the oxidized form. In the dark the dye
again becomes colorless. The initial rate
of this photochemical reaction has been
used as a measure of system I action by
Brown {Year Book 70, pp. 499-504,
1971), and also by Binder, Tel-Or, and
Avron (1976). The measurement of the
equilibrium dye concentration rather
than its photooxidation rate is, however,
far more convenient and precise.

The reaction mixture for DCIP oxi-
dation was: MgCU 10-^ M; DCIP 6.6
X 10~-^ M; bovine serum albumin 0.04% ;
DCMU 10-^ M; Methyl viologen 20
/xM; Tris pB. 7.42, 0.05 M; sucrose 0.5
M; green material 4.6 /xg chl/ml (56%
absorption at 680 nm) ; catalase 1.66 X
10-2 mg/ml; sodium ascorbate (to re-
duce dye + 10% excess) ; dissolved O2
about 2.8 X 10"^ M. DCMU was used
to poison system II activity, methyl
viologen to couple the reduced dye to
dissolved oxygen, and (at Dr. Avron's
suggestion) catalase to decompose any
H2O2 produced. Bovine serum albumin
and sucrose were added as stabilizers. At

DEPARTMENT OF PLANT BIOLOGY

215

first, enough ascorbate was put in to re-
duce the DCIP and to leave an excess of
about 10%. Later, more ascorbate was
usually added to keep the reduction rate
at a convenient value.

Calculations

During a run of several hours with
20-25 light exposures, the amount of dye
oxidized by a standard exposure in-
creased about 20% because of ascorbate
depletion. The amount of dye oxidized at
the dynamic equilibrium during the ex-
posure was not linear with light intensity.
Therefore, a single run, as with the DCIP
reduction experiments, consisted of fre-
quent exposures to (1) an intermediate
intensity of a standard wavelength, 680
nm; (2) a series of various intensities of
this wavelength; and (3) a series of
wavelengths adjusted in intensity to give
responses within the range of the re-
sponses to the standard wavelength. The

S3

2 -

"I 1 ' r

^ ■V H ti*^

iH*t^

dye oxidation level was monitored by
light transmission at 593 nm iTig. 3j.
The transmission change caused by ex-
posure to the standard intensity at vari-
ous times was used to correct all responses
to a value equivalent to that at the start
of the run. The corrected deflections for
the standard wavelength of 680 nm were
then plotted against intensity. From this
curve (Fig. 4) the intensity of 680 nm
needed to give a response equal to that
obtained from each wavelength was
determined.

For the dye reduction experiments a
computer program, ACTI5, was used to
evaluate the action per unit of average
actinic intensity in the vessel for all the
exposures to the standard wavelength and
to the experimental wavelengths. All
calibration factors were included in the
program. With the absorption spectrum
of the green material for each run and
with the dye absorption spectrum, it was
possible to calculate from the measured
transmission the average light intensity
within the sample. The transmission was
used to determine the dye concentration

< 2

12 3 4

Time, min

Fig. 3. Nostoc particles suspended in solutions
of DCIP kept reduced to the colorless form by
excess ascorbate will photooxidize some of the
dye until a steady state is reached at a low
concentration of oxidized dye. In this case the
light effect is measured by the difference in
level in the light and dark. Here extrapolation
is not necessary because the dye returns to the
colorless reduced state after the exposure. The
light intensity is measured by the upper trace.

Intensity, rel quanta min

Fig. 4. The transmission change for various
intensities of 680-nm hght. This curve is used
to determine the intensity of 680 nm equivalent
in action to the exposures for different wave-
lengths. Very similar curves relate the rate of
dye reduction to hght intensity.

216

CARNEGIE INSTITUTION

at the middle of each light exposure. The
response to a standard exposure decreased
with time for the dye reduction experi-
ments. Corrections for this effect were
made either graphically or by use of the
polynomial curve fitting program POLYR
adapted to our computer by Ford and
Brain. Similarly, either that program or
graphical interpolation was used (plotting
response versus intensity, as in Fig. 4) to
calculate the intensity of the standard
wavelength required to give the same
response as that obtained from each ex-
perimental wavelength. Thus completely
comparable action responses were ob-
tained for each wavelength within a
single run of several hours.

The action points were plotted against
the fractional absorption of the green
material used for each run. The action
points were scaled to three fourths the
height of the fractional absorption curve
at its peak. This arbitrary scaling factor

was a compromise between the theory
that supposes all pigments to be active
and another theory that assumes half
the absorption at the peak to be inactive.
Fortunately, the factor chosen has only a
small effect on the final spectrum. The
action points for each run were then
transformed to active optical density by
the relation, E' = logio (1/[1-A']).
These tables of active optical density
may then be scaled relative to each other
without regard to the concentration of
pigment in the separate experiments. It
is important to note that the reasons for
using active optical density rather than
fractional absorbance in presenting ab-
sorption spectra apply equally well to
action spectra (French, 1977).

Results

The advantage of the steady-state
measurements used in the photo-oxida-

Q
O

>

u
<

u

c
o

_Q

_Q
<

Numb

er

of poinfs for DC IP reduction:

2

3

21 I52I3I525756

5

4

8 7

3 1

I.O

-

/

^

V

-

0.9

-

//'

■U

\

-

0.8

-

/

\

A

-

0.7

-

Absorption A

o DCIP oxidation /'

\

< 1
J

\

-

0.6

-

—\- DCIP reduction /'

\

\

-

0.5

-

III
//
/ /

» \
\ \

«r \

-

0.4

-

1 i

\

1 \
\ \
\ \

-

0.3

-

y^^^ ^^^- ^ '

\
\

\

1 1
\
\

\

y

0.2
0.1

T

4_

r

o

f^r\. .n silt half-widtti = 4nm

>H

o\

XV -

r\

1

1 1 1 1 1 1 1

1

1

1 1

600

650
Wavelength, nm

700

Fig. 5. The action .spectra for DCIP reduction with bars showing the standard deviation. The
number of mea.suremont.s is indicated above. Points for DCIP oxidation are shown as circles.
The ab.'iorption ."Spectrum of one preparation is given as a hne. The inactive absorption at
shorter wavelengths is presumably due to some phycocyanin in the preparation.

DEPARTMENT OF PLANT BIOLOGY

217

tion reaction can be seen by comparing slightly improved the fit of the action
Figs. 2 and 3. In Fig. 3 the base line curve to the absorption. It therefore
from which the light effect is measured is seems that the DCIP oxidation photo-
located by the line through the transmis- reaction uses both pigment systems I and
sion curves position both before and after II for the reaction and that system II is
the exposure. The response can be easily perhaps slightly more effective than pig-
evaluated from such data. ment system I for this reaction.

For the reduction reaction, however, The deviation of both action spectra

many averaged repetitions were needed.
Figure 5 shows the number of determina-
tions and their standard deviation as bars

from the absorption curve in the 600-650
region may have been caused by small
amounts of inactive phycocyanin re-

for each point on the DCIP reduction maining in the preparation, even though

action spectrum. Superimposed is the it was more thoroughly purified than

action spectrum for dye oxidation (cir- earlier preparations. The deviation of the

cles), and the absorption spectrum for action spectra from the absorption be-

one of the preparations is given as a line, yond 695 nm indicates some inactive ab-

This unexpectedly close agreement may sorption in this region. The expected dif-

be the result of MgCU facilitating energy ference on the long-wavelength side of

transfer between pigments, possibly the peaks between what are supposed to

through aggregation, so that the entire be typical system I and system II actions

pigment system functions as a single are in the direction usually found in

assembly. The differences in the two green plants. The great overlap between

action spectra in the 600-650 region may the two actions is, however, not typical,
be significant. The curve for DCIP re-
duction appears to peak near 620-625,
while the curve for its oxidation seems
higher, at 625-635.

The DCIP oxidation action spectrum
nearly fits the entire absorption spectrum.

References

Binder, A., E. Tel-Or, and M. Avron,
Eur. J. Biochem. 67, 187-196, 1976.

Attempts to improve the fit by adding French, C. S., Photochem. and Photobiol.

the two action spectra were not success- 25, 159-160, 1977.

ful. Subtraction of about 7% of the Vernon, L. P., and W. S. Zaugg, J. Biol.

DCIP reduction action curve, however, Chem. 235, 2728-2733, 1960.

PHOTOSYNTHETIC ELECTRON FLOW FROM
PHOTOSYSTEM II REQUIRES MEMBRANE-BOUND

BICARBONATE

Alan Stemler

Bicarbonate ions, which are required
for the photochemistry associated with
photosystem II, are bound to chloroplast
thylakoid membranes in two distinct
pools (Stemler, Year Book 75, pp. 477-
479; 1977). One site has a relatively high
affinity for HCOs" and exists in a con-
centration approximating that of the
photosystem II reaction centers. More
numerous lower-affinity sites also exist.

Of the two binding sites, the high-
affinity site is easier to characterize. Once
HCOs" is bound to this site, it cannot be
removed by repeated washes with buf-
fered solutions at neutral pH. The amount
of HCOa" bound to the large pool, on
the other hand, diminishes with each
washing, though trace amounts persist
even after three washes. Washing away a
large fraction of the HCOs" bound to

218

CARNEGIE INSTITUTION

TABLE 1. Ability of Various Wash Media to Remove Bound H^COs" and Suppress

Oxygen Evolution *

Wash solution

HCO.-
Removed
dpm-' Chi (% of
X 10"' control)

pinoles O2 mg ' Chi hr "'

+HCO3-
-HCO3- +HCO3- -HCO3-

0.1 M sodium phosphate pH 7.0

0.1 M XaCl 440

0.3 M sucrose

0.1 M sodium phosphate pH 5.0

0.175 M XaCl 3

0.1 M sodium formate

lO-* M DCMU (added first)

0.1 M sodium phosphate pH 5.0 249

0.175 M XaCl

0.1 M sodium formate

control

99

44

40.0

2.6

41.3

38.7

1.03

14.9

* The chloroplasts were charged with IP^COg" by suspension in 1 ml medium containing O.IM
sodium phosphate pH 5.8, 0.2 M NaCl, 168 fiM NaH'^COs (1.2 ^ Ci) and 2 mg chl. After a
5-min incubation at 30°C, 6 ml of ice-cold wash medium containing O.lAf sodium phosphate
pH 7.0, 0.01 M XaCl, 0.3 M sucrose was added. After centrifugation the pellet was washed
twice in 7 ml of the wash medium, then a third time in the medium indicated. The dpm were
determined in the final pellet and portions were measured for oxygen-evolving ability in
saturating light immediately after suspension in reaction mixture containing 0.1 M sodium
phosphate pH 7, 0.175 M NaCl, 0.1 M sodium formate, 2 mM K3Fe(CN)6, 50 ^g Chl m\-\
Initial rates were measured before and after injection of NaHCOs to 12.5 mM.

low-affinity sites does not induce de-
pendence of oxygen evolution on added
HC03~. This result indicates either that
the large pool of binding sites is not in-
volved at all in photochemical events or
that only a very small fraction of the
low-affinity sites need to be occupied with
HCOo" at any given time.

The role of HCOa" bound in the small
high-affinity pool is more evident. Chlo-
ropla.-t membranes were depleted of
HCO-i" (see Stemler and Radmer, 1975,
for details), then provided with enough
i^C-labeled HCO," to refill the small
pool. They were then washed several
times in buffered solution, pH 7, to re-
move all ''trapped" Hi^CO.-. Such chlo-
ropla.sts require no added HCO.s" for full
oxygen-evolving activity (Table 1, top
line). If such chloroplast membranes are
washed once in "HCO.^" depletion me-
dium" containing high salt concentra-
tions and buffered at pH 5, all the bound
H^'^COa" is removed and oxygen evolving

capability is reduced to a trace (Table
1, line 2). Adding back HCOa" then
stimulates oxygen evolution 15-fold. This
result shows clearly that the depletion
procedure does in fact remove membrane-
bound HCOs" and that this bound anion
is needed for maximum photosystem II
activity. If the inhibitor 3-(3,4-dichlo-
rophenyl)-l, 1-dimethylurea (DCMU)
is given to the chloroplasts before they
are suspended in depletion medium, the
removal of bound HCOs" is retarded
(Table 1, line 3). This inhibition sug-
gests that the DCMU binding site and
the HCOs" binding site are very close.
The effect of light on high-affinity
bound HCOs" is of interest as a clue to
the mechanism of action of this ion.
Chloroplast grana were labeled with
H^'^COa", then washed several times to
remove "trapped" H^^COs". They were
then illuminated in the presence of ferri-
cyanide while oxygen evolution was mon-
itored. At various times, aliquots of the

DEPARTMENT OF PLANT BIOLOGY

219

0.5

8 12

Time of illuminafion, min

- I

I

o

X

3 2

2 e

a

Fig. 6. Loss of tightly bound H^COa from chloroplasts during oxygen evolution. Chloroplasts
were charged with H^COs" as explained in the legend of Table 1. After a third washing the
grana were suspended in reaction mixture containing 0.1 M sodium phosphate pH 7.0, 0.01 M
NaCl, 0.3 M sucrose, 0.01 M K3Fe(CN)6, 162 ^g chl ml"\ then illuminated. At times indicated,
1 ml of the reaction suspension was withdrawn from the illumination chamber, centrifuged, and
the pellet measured for bound HCOs".

cyanide while oxygen evolution was mon-
itored. At various times, aliquots of the
reaction mixture were withdrawn and
centrifuged; the grana-containing pellet
was then assayed for bound H^^COs".
Results are shown in Fig. 6. Chloroplasts
that were illuminated for 18 minutes
gave off oxygen continuously, though at
a decreasing rate. At the same time only
about a third of the bound HCOs" was
lost from the thylakoid membranes. The
same result was observed when unlabeled
HCOa" was added to the reaction mix-
ture until a final concentration of 20 mM
was reached. This result indicates that
bound HCOs" does not exchange with
free HCOs" in the course of oxygen evo-
lution. Instead, the single ion bound to
the photosystem II reaction center com-
plex remains in place while hundreds of
molecules of oxygen are evolved. The
fraction of bound HCOs" which was lost

during illumination may reflect non-
specific membrane damage. Another pos-
sibility is that the lost HCOs" represents
HCOs" that was bound in the low-
affinity pool and managed to survive the
wash treatment used before illumination
to remove trapped HCOs". If the second
possibility holds true, it would imply that
the low-affinity HCOa" binding sites may
also play a role in the photosystem II
reaction.

Further characterization of the two
pools of bound HCOs" is in progress.

References

Stemler, A., Biochim. Biophys. Acta 460,

511-522, 1977.
Stemler, A. and R. Radmer, Science 190,

457-458, 1975.

220

CARNEGIE INSTITUTION

THE EFFECT OF TEMPERATURE ON THE PHYSICAL PHASE

OF CHLOROPLAST MEMBRANE LIPIDS AND

PHOTOSYNTHESIS

David C. Fork, Norio Murata, and Mordhay Avron

Introduction

These studies and those reported in the
hist two Year Books seek a better un-
derstanding on the molecular level of
the influence of temperature on photo-
synthesis.

The photosynthetic apparatus that
converts solar energy to chemical energy
is located in lamellar lipoprotein struc-
tures that constitute the chloroplast thy-
lakoid membrane. One model (for a
re\iew, see Anderson, 1975) views the
chloroplast thylakoid membrane as a
lipid bilayer containing embedded pro-
teins and certain molecules involved in
the bioenergetic process such as the chlo-
rophylls, quinones, various cytochromes,
and carotenoids. Membrane lipids can be
tightly bound to the outer layers of
embedded proteins or can act with one
another to form a fluid bilayer through
which the lipid molecules can move
laterally (Jost et al., 1973; Tralible and
Overath, 1973; Dehlinger et al, 1974).

In earlier studies we found that at the
temperature of transition of the physical
phase of membrane lipids from the liquid
crystalline to the phase-separation state
there appeared characteristic changes of
photosynthetic electron transport {Year
Book 74, 766-776; Murata et al, 1975;
Fork and Murata, 1977).

In the present studies we report phase
changes of thylakoid membrane lipids in
chilling-resistant plants such as spinach
at sub-zero temperatures and, by con-
trast, in the extreme thermophilic blue-
green alga Synechococcus lividus where
phase changes were seen at the highest
temperatures f43''C) so far observed for
a photosynthetic organism.

In studies of the temperature depend-

ence of the light-induced spectral shift
in carotenoids and of proton uptake and
efflux using the fluorescent indicator 9-
aminoacridine, we found that diffusion
of ions through thylakoid membranes is
greatly affected by the physical phase of
membrane lipids and that the membrane
becomes leaky to the ions below the phase
transition temperature.

Lipid Phase Changes in Lettuce and

Spinach Chloroplasts at Sub-Zero

Temperatures

Norio Murata and David C. Fork

We have shown in our previous studies
that fluorescence is a convenient tool to
detect the transition of the physical phase
of thylakoid membrane lipids in photo-
synthetic material (Murata and Fork,
1975; Murata et al, 1975; Fork and
Murata, 1977). A maximum or positive
shoulder in the fluorescence-versus-tem-
perature curve indicates the occurrence
of a phase transition.

Except for the desert plant Tidestromia
oblongifolia (Murata and Fork, 1977) no
maxima were found in higher plant chlo-
roplasts above 0°C (Murata and Fork,
1975). In this study we present the
fluorescence-versus-temperature curves
for lettuce and spinach chloroplasts from
about 5°C to — 35°C.

To prevent freezing of the reaction
mixture we added ethylene glycol to
make a concentration of 50%. This con-
centration partly inhibits the Hill re-
action using 2,6-dichlorophenol indo-
phenol (Inoue and Nishimura, 1971) and
will protect against freezing down to
about -35°C (Cox, 1975).

Figure 7 shows the temperature de-

DEPARTMENT OF PLANT BIOLOGY

221

100

U 90

80 -

-31° -25°
I I

,(c) + 100 mM KCI

/"

(b) + 5 mM MgCl2

(a) + 5 mM NaCI

-50 -40 -30 -20 -10

TEMPERATURE, °C

10

Fig. 7. Temperature dependence of chloro-
phyll a fluorescence in chloroplasts of lettuce
grown at 25 °C, (a) The suspension medium
consisted of a mixture of ethylene glycol and
water (50:50 v/v) that contained 200 mM
sucrose, 5 mM NaCl, 1 mM sodium ascorbate,
5 X 10"' M DCMU, and 5 mM Tricine-NaOH
buffer, pH 7.4, (b) plus 5 mM MgCL, (c) plus
100 mM KCI. Excitation light: 435 nm, 500 ergs
cm"^ sec"^. The fluorescence was measured at
685 nm. Temperature was lowered at a rate of
rC/min.

pendence of steady-state chlorophyll a
fluorescence in chloroplasts isolated from
lettuce grown at 25°C. At a low mono-
valent cation concentration (5 mM Na + )
the fluorescence maximum appeared at
— 11°C. Addition of 5 mM Mg++ or 100
mM K+ lowered the maximum tempera-
ture to — 31C° or — 25 °C, respectively.

1 1

-31°

1

1 1

100

—

1

>»

—

-20°

1

V

(b) + 5 mM MgCl2

90
an

1 1

(a) + 5 mM NaCI
1 1

-40 -30 -20 -10

TEMPERATURE, °C

10

Fig. 8. Temperature dependence of chloro-
phyll a fluorescence in spinach chloroplasts in
the presence of 5 mM NaCI or 5 mM MgCh.
The experimental conditions were the same as
in Fig. 7.

Figure 8 shows the temperature de-
pendence of chloroi)hyll a fluorescence in
spinach chloroplasts. In the presence of
5 mM Na+ the maximum appeared at
— 20^0 and in the presence of 5 mM
Mg++ it appeared at — SrC.

A maximum in the temperature-versus-
fluorescence curve indicates a transition
of the physical phase of thylakoid mem-
brane lipids. The curves obtained using
lettuce or spinach chloroplasts are very
broad when compared to those obtained
from Anacystis nidulans (Murata and
Fork, 1975; Murata et al, 1975) or for
Synechococcus lividus (next article), for
example. This indicates that the phase
transition of membrane lipids in higher
plants occurs over a broad range of tem-
peratures. In addition, the temperature
of the phase transition and perhaps mem-
brane fluidity are greatly influenced by
the presence of divalent cations and by
higher concentrations of monovalent
cations, at least in higher plants.

We used thylakoid membrane prepara-
tions of Anabaena variabilis and an arti-
ficial membrane preparation of dimyris-
toyl lecithin to see if the addition of
50% ethylene glycol had an influence on
the phase transition of membrane lipids
under conditions in w^hich the phase
transition occurred above freezing tem-
peratures. Figure 9 shows that the maxi-
mum in the fluorescence-to-temperature
curve appeared at 10°C either with or
without ethylene glycol added. This con-
centration of ethylene glycol increased
the absolute fluorescence yield by about
20%, but had no effect on the phase
transition between the liquid-crystalline
and the phase-separation states.

In a water suspension of dimyristoyl-
phosphatidylcholine the fluorescence yield
of l-anilinonaphthalene-8-sulfonic acid
(ANS) decreased drastically at 23°C
with and without ethylene glycol. The
same temperature dependences were ob-
served when the temperature was de-
creased or increased. This dramatic
fluorescence change is known to occur at
the phase transition between the liquid-

222

CARNEGIE INSTITUTION

10°

I 1

100

y

1

50% EG
50% WATER

90

1

^

1

WATER
1 1

-

-10

10 20 30

TEMPERATURE °C

40

Fig. 9. The effect of ethylene glycol on the
temperature curve of chlorophyll a fluorescence
in a thylakoid membrane preparation of
Anahaena variabilis (strain M3) that was grown
at 33 'C. The fluorescence yields for each curve
were measured using the same sensitivities. The
suspension medium with and without 50%
ethylene glycol contained 200 mM sucrose, 5
mM XaCl, 1 m.V sodium ascorbate, 5 X 10'^
M DCMU, and 5 mM Tricine-XaOH buffer,
pH 7.2. The experimental conditions were as
described for Fig. 7.

crystalline and the gel states (Traiible,
1971: Oldfield and Chapman, 1972), and
it appears that 50% ethylene glycol has
no effect on this change. The absolute
fluorescence yield decreased to about one
fourth after addition of ethylene glycol.
This decrease may be caused by a change
in the partitioning of ANS between the
membrane lipid and the water phase
when the aqueous phase contains ethyl-
ene glycol.

It is clear that 5 mM Mg++ and 100
mM K^ shifted the phase transition to
lower temperatures. This suggests that
the phase transition and perhaps mem-
brane fluidity are markedly affected by
these cations. Cations exert striking
effects on lipid phase transitions in model
membranes (van Dijck et al, 1975;
Onishi and Itoh, 1974). In this case,
however, the cations produce effects
opposite from those seen with chloro-
plasts. In may be that cations interact in
thylakoid membranes with negatively

charged lipids such as sulfoquinovosyldi-
glyccride, phosphatidylglycerol and phos-
phatidylinositol, and so affect the cluster-
ing of these lipids, thereby influencing the
transition of the physical phase of mem-
brane lipids.

Studies on the Effect of Transition

OF THE Physical Phase of Membrane

Lipids on Electron Transport in the

Extreme Thermophile

Synechococcus lividus

David C. Fork and Norio Murata

As mentioned in the previous article,
the fluorescence of chlorophyll a in vivo
is a convenient probe to detect transitions
of the physical phase of thylakoid mem-
brane lipids in photosynthetic material.
Using this technique, we measured for a
number of photosynthetic organisms
phase transition temperatures ranging
from below zero for spinach or lettuce
chloroplasts (previous article) to around
24°C in the blue-green alga Anacystis
nidulans grown at 38 °C (Murata and
Fork, 1975; Fork and Murata, 1977;
Year Book 75, 465-472). We used the
blue-green alga Synechococcus lividus,
w^hich is capable of growing at the high-
est temperature (73-75°C) of any photo-
synthetic organism known (Kempner,
1963; Brock, 1967; Castenholz, 1969).
This extreme thermophile is found in
alkaline hot springs where it often grows
abundantly.

We measured phase transitions for
these algae grown at various tempera-
tures and found significant alterations in
the rates of photosynthetic electron trans-
port above and below these phase transi-
tion temperatures.

The cultures of Synechococcus lividus
used for these experiments were obtained
through the courtesy of Dr. Richard W.
Castenholz of the Biology Department
of the University of Oregon and Dr.
Mercedes R. Edwards of the New York
State Department of Health. All of the
cultures were grown in D-medium
(Castenholz, 1969) at the desired tern-

DEPARTMENT OF PLANT BIOLOGY

223

perature with a gas phase of 0.5% CO2/
air in bubble tubes fitted with condensers
to retard evaporation. The intensity of
the tungsten Hght used for growth was
about 5000 lux. The measurements of
fluorescence versus temperature were per-
formed as described previously (Murata
and Fork, 1975). Electron transport was
followed by measuring redox changes of
cytochrome / at 420 nm. The outputs of
the photomultiplier that measured the
absorption change, of the thermocouple
that measured temperature in the cuvette,
and of the photocell that monitored light
intensity were all amplified and trans-
mitted to an analog-to-digital converter
that digitized the signals at intervals

appropriate to tlic needs of the experi-
ment. The digitized values were then
stored and analyzed by a Hewlett-
Packard 2116 computer such that the
desired on or off rates or the steady-state
values of absorption changes, tempera-
ture, and light intensity could all be
obtained and displayed during an experi-
ment in which temperature was decreased
or increased at rates of about 1°C per
minute.

Measurements of the fluorescence of
chlorophyll a as a function of tempera-
ture were all done in the presence of 10
ixM DCMU [3-(3,4-dichlorophenyl)-
1,1-dimethylurea] to inhibit electron
transport. Figure 10 shows the tempera-

— I 1 1 1 1 1 r

Synechococcus lividus

Grown af 65 °C

45 °C

Temperature, °C

Fig. 10. Temperature dependence of chlorophyll a fluorescence in Synechococcus lividus
grown at 65°, 60°, 55°, and 45°C. Fluorescence was excited at 435 nm (400 ergs cm~- sec~^) and
measured at 685 nm in the presence of 10 fiM DCMU as described previously {Year Book 75,
pp. 465-^72). The cells grown at 65°C were strain OH68-S, Clone HXf (Meeks and Castenholz,
1971). Those grown at 60°, 55°, and 45°C were strains SY-2, SY-4 and SY-3, respectively,
obtained from Dr. M. Edwards.

224

CARNEGIE INSTITUTION

ture-versus-fluorescence curves for cells
of Synechococcus lividus. The maxima,
or shoulders, in these curves mdieate
transitions of the physical phase of
thylakoid membrane lipids. It can be
seen that upon decreasing and then in-
creasing the temperature of cells grown
at 65 ^\ 60 "\ and 55 °C the maximum or
shoulder occurred at about 43^C. The
maximum could be seen for several cycles
of temperature increase and decrease.
Cells grown at 45^C had broad maxima
in their temperature-to-fluorescence
curves ranging from 18° to 23°C. Upon
increasing temperature a second shoulder
or maximum appeared at 53° and 45 °C
in cells grown at 55° and 45°C respec-
tively. These maxima do not correspond
to a lipid phase transition, since they
were not observed on decreasing the
temperature.

In the photosynthetic electron trans-
port system of a blue-green alga like
Synechococcus the electron carriers such
as plastoquinone, cytochrome /, and P700

are oxidized upon excitation of system I
and reduced by illumination in pigment
system II (Amesz and Duysens, 1962;
Amesz, 1964; Murata and Takamiya,
1969). To observe the effect-of the transi-
tion of the phase of membrane lipids on
electron transport reactions, the oxida-
tion-reduction reactions of cytochrome /
were measured at various temperatures.
Cells were illuminated with high-
intensity red actinic light having wave-
lengths longer than 620 nm; the light
was absorbed by both pigment systems.
Absorbance changes produced by cyto-
chrome / (and partly by P700) were
measured at 418-420 nm near the Soret
maximum for cytochrome /. Light-minus-
dark difference spectra at 55°C showed
the characteristic shape produced by
light-induced oxidation of cytochrome /
and had a maximum near 405 nm and
minima at 422 and 554 nm. A shoulder
around 433 nm, produced by P700, was
also visible.
Figure 11 shows examples of the

e

c
00

0)
u

c
o

JQ

JQ
<

Fig. 11. Kinetics of light-induced ab.sorbance changes at 418 nm in cells of Synechococcufi
lividu.s grown at 55''C and measured at 53% 34°, and 22''C. Half-bandwidth of the measuring
beam was 2 nm. Red actinic liglit with wavelengths from 620 to 750 nm was obtained with glass
optical filters, Schott RG2 and Calflex C, and had an intensity of 1.8 X 10' ergs cm'" sec"'.

DEPARTMENT OF PLANT BIOLOGY

225

kinetics of absorbance changes at 418 nm
measured at 53°, 34°, and 22 °C for
Synechococcus cells grown at 55°C. Upon
illumination with bright red light, a
rapid oxidation (absorbance decrease)
was seen. This was followed by a rapid
transient reduction of the oxidized cyto-
chrome in the light and then by a second
and slower oxidation that reached a
steady-state level during continued illu-
mination. Prompt reduction of cyto-
chrome / took place when the actinic
illumination was turned off. Electrons
originating from the reduced plasto-
quinone produced by exciting system II
with high-intensity red light appear to be
responsible for the transient reduction of
cytochrome / observed during illumina-
tion, since this transient reduction was
inhibited by DCMU and was not seen
with far-red actinic light, which excites
only system I. Both the transient reduc-
tion of cytochrome / in the light and its
reduction in the dark were temperature-
sensitive (Fig. 11).

50

Tempera+ure, C

40 30

c
.9

u

o

100

30

10

o - o
4.6 kcal/mol

8.4 kcal/mol

3.1

3.2

3.3

3 0,y-[

l/T X 10 . "K

Fig. 12 Arrhenius plot of cytochrome / reduc-
tion measured at 420 nm after turning off red
actinic light (described in Fig. 11) in intact cells
of Synechococcus lividus grown at 55°C. Light
and dark periods used were 16 and 4.3 sec,
respectively. Temperature was lowered from
51° to 33° C at a rate of about TC/min.

Figure 12 shows the temperature de-
pendence of the dark reduction of cyto-
chrome / after cells grown at ^o(^ wore
illuminated with red actinic light. The
graph gives a fairly straight line with
an abrupt change in slope at 43'^C, near
the phase transition temperature. The
activation energies above and below the
break point are 4.6 and 8.4 kcal/mole,
respectively. Meeks and Castenholz
(1971) found a break near 43 "^C in the
Arrhenius plots for photosynthesis meas-
ured as photoincorporation of ^^C-
NaHCOs by Synechococcus lividus cul-
tures grown at temperatures above 60°C.

The temperature dependence for the
transient reduction of cytochrome / dur-
ing the light exposure was measured. The
time required in the light for the transient
to change to one fourth of its initial
extent was set as the standard for the
relative rate of electron transport from
plastoquinone to cytochrome /. A single
sample of cells grown at 55 °C was first
cooled from 54°C to about 33°C and
then heated again to about 57°C; and
relative rates of cytochrome / reduction
in the light were determined. The
Arrhenius plot (Fig. 13) shows that the
points clustered along two straight lines
that intersected at 43 °C — the phase
transition temperature as determined
by the temperature-versus-fluorescence
measurement. The activation energies
above and below the break point fol-
lowed each other with about the same
activation energies. Above the break
point, however, the slopes of the lines
were lower than before (2.3 kcal mol~^
on initial cooling; —3.4 kcal mol~^ upon
rewarming). This decrease may result
from damage to the thylakoid mem-
brane when the temperature was low-
ered. The ''chilling" of the cells to 33°C
may have resulted in a lower rate of
electron transport when the temperature
was raised again. A similar result was
obtained (Murata et a/., 1975) with
Anacystis grown at 38°C, cooled to 6°C,
and heated again to 33 °C. A second
break occurs at 53 °C near the growth

226

CARNEGIE INSTITUnON

Temperature, C

SO

30 n r-

3.3

l/T X 10^, °K"

Fig. 13. Arrheniiis plot of the transient reduction of cytochrome / measured at 420 nm
during a period of red actinic ilhimination (described in Fig. 11) in intact cells of Synechococ-
cus Uvidus gro\\Ti at 55° C. The time for the transient seen immediately after illumination to
decay to one fourth of its initial value was taken as the relative rate of electron transfer from
photosA'stem II to cytochrome /. One sample of cells was used for the experiment and was first
cooled and then heated at rates of about l°C/min. Cycles of 0.3 sec light and 10 sec dark
were used.

temperature for the cells. Above this
temperature the rates drop significantly.

Transitions of the physical phase of
the membrane lipids produce significant
alterations in photosynthetic electron
transport (Shneyour et al, 1973; Murata
et al, 1975; Fork and Murata, 1977), in
the state 1 to state 2 shift (Murata et al.,
1975), in the intensity of delayed fluores-
cence (Ono and Murata, 1977) , in mem-
brane permeability to nonelectrolytes
(Nobel, 1974), and in the light-induced
production of a membrane potential (fol-
lowing articles). The change in the rate
of cytochrome / reduction coresponding
to a transition of the phase of membrane
lipids suggests that, as with Anacystis
(Murata et al., 1975), the part of photo-
synthetic electron transport involving
participation of the lipophilic molecules
of plastoquinone is significantly affected
by change.- in the phase of thylakoid
membrane lipids.

The.se studies indicate that the preser-
vation in the thylakoid membranes of
the proper physical state of the lipids is

a requirement for efficient photosynthetic
function and that many organisms are
flexible and able to achieve this state
even if their growth temperature is
changed.

Acknowledgments

We are indebted to Mr. Glenn A, Ford,
who wrote the programs and helped in-
terface the apparatus to the computer
system, and to Mr. Benny Catanzaro,
who designed and built electronic com-
ponents used in these experiments.

The Light-Induced Carotenoid Shift

IN Cyanidium and in Higher Plant

Leaves as an Indicator of Phase

Changes in Chloroplast

Membrane Lipids

Norio Murata and David C. Fork

In this study we investigated the tem-
perature dependence of the light-induced
spectral shift of carotenoids in an at-
tempt to demonstrate an effect of the

DEPARTMENT OF PLANT BIOLOGY

227

physical phase of membrane lipids on the
production of a membrane potential and
on the permeability of the membrane to
ions.

The absorbance changes that are seen
near 520 nm (positive) and 475 nm
(negative) upon illumination of higher
plants and green algae appear to be pro-
duced by a spectral shift of photosyn-
thetic pigments (mainly ^-carotene) in

the thylakoid membranes (Pv,eich ('A al.,
1976; Kleuser and Biichen, 1969; Amesz
and Visser, 1971). Light-induced absorb-
ance changes with three maxima and
minima, principally due to a spectral
shift of carotenoids, have been observed
in several classes of algae. Similar ab-
sorbance changes are also seen in isolated
higher plant chloroplasts (Weikard et aL,
1963; Wakamatsu and Nishimura, 1974 j.

O

O

C

SI

l_

o

_Q
<

T 1 1 1 1 r

405

"T I I 1 1 r-

Measured at 31 °C

400

450 500

Wavelength, nm

550

Fig, 14. Light-minus-dark difference spectra measured at 31° and 9°C in intact cells of
Cyanidium caldarium grown in Allen's (1959) medium at 38° C. The half -bandwidth of the
measuring beam was 2 nm. Red actinic light with wavelengths from 630 to 750 nm was obtained
by filtering white light through optical filters. Filters used were Schott RG2, 3 mm and Calflex
C and 37 mm of water. The light had an intensity of 1.8 X 10^ ergs cm"" sec"^. Far-red light
was obtained in the same way after substituting Corning glass CS 2-64 for RG2 and had an
intensity of 1.3 X 10^ ergs cm"^ sec"^ in a band from 650 to 750 nm. The absorbance difference
was between the levels in the light after 1 sec of actinic illumination and the level in the dark
1.6 sec after turning off the actinic illumination.

228

CARNEGIE INSTITUTION

Although the unicellidar alga Cya-
nidium caldarium resembles a blue-green
alga in its preponderance of the chromo-
protein pigment c-phycocyanin. its mor-
phological and cytoplasmic organization
is more like that of a red alga (Doemel
and Brock. 1971 K Figure 14 shows that
the light-minus-dark difference spectra
measured for this organism are not unlike
those obtained using other red algae
(Fork and Amesz. 1967), since they have
positive peaks at 450, 483, and 515 nm
and negative peaks at 433, 465, and 500
nm. These reversible alternating positive
and negative peaks, each separated by
about 33 nm, can be attributed to a light-
induced shift of the absorption spectrum
of carotenoids to longer wavelengths in
the light, followed by a reversal in the
dark. The difference spectra of Fig. 14
also show a reversible light-induced
oxidation of cytochrome /, with negative
peaks at 554 and 420 nm and a positive
peak near 405 nm.

At low temperatures the dark decay of
the absorbance change produced by the
carotenoid shift was composed of both a
fast and a slow component. The decay
rates of the slow component were plotted
in Fig. 15. In this experiment the recipro-
cals of the half-decay time were taken
as representing the decay rates even
though the dark decay did not follow
perfect first-order kinetics. The Arrhenius
plots followed essentially the same line
upon decreasing or increasing the tem-
perature. A break in the line appeared
at 8''C. This temperature corresponds to
the temperature of phase transition in
Cyanidium, using chlorophyll a fluores-
cence measurements ( Year Book 75, 465-
472 1. The apparent activation energies
above and below the 8°C break point
were 11 and 40 kcal/mole, respectively.

The desert plant Tidestromia oblongi-
John grows at high temperatures (^
50^C) in Death Valley, California {Year
Book 70^ 540-550) . For measurements of
fluorescence versus temperature we used
chloroplasts extracted from Tidestromia
collected one summer morning in Death

Tempera+ure, C

40

30

20

10

30

4

\

1

^

1

10

-tY

A^

\

3

1

1

1

1

1 1

3.2

3.4

3.6

3 0,y-\

l/T XIO'', °K

Fig. 15. Arrhenius plot of the decay kinetics
of the slow component of the light-induced
carotenoid change measured at 483 nm in C.
caldarium. Reciprocals of the half-decay times
were taken as representing decay rates. Open
circles indicate decreasing temperature; open
triangles, increasing temperature. Red actinic
light was used, as described in Fig. 14. Re-
peated hght and dark periods were 4.0 and 4.2
sec, respectively.

Valley and used the same afternoon. With
this preparation we found a shoulder or
in some cases a maximum around 5°C
in the fluorescence-versus-temperature
curve (data not shown). Previously
(Murata and Fork, 1975; Year Book 7l,
766-776) we had not seen such a break
in the fluorescence-to-temperature curve,
possibly because the plants used in those
experiments were cultivated in a growth
cabinet at temperatures below those
found in the native environment.

Figure 16 shows the Arrhenius plot
of the dark decay of the slow component
of the 520-nm change (carotenoid change)
in a leaf of T . oblongifolia. The points
obtained upon decreasing and increasing
the temperature followed the same line.
At 5°C a clear break in the line was

DEPARTMENT OF PLANT BIOLOGY

229

Temperafure, C
30 20 10

l/T XIO^ °K

Fig. 16. Arrhenius plot of the decay kinetics
of the slow component of the absorbance change
at 520 nm in a leaf of Tidestromia oblongifolia.
The plant used for these measurements was
collected during the summer growing season in
Death Valley, California. Reciprocals of the
half-decay time were taken as representing the
decay rates. Open circles are used for decreasing
temperature ; open triangles, for increasing
temperature. Repeated cycles of 5.1 sec of red
light (described in Figure 14) and 7.4 sec dark-
ness were used.

seen. Thi.s temperature corresponds to
the phase-transition temperature de-
tected by the fluorescence measurement
mentioned above. The apparent activa-
tion energies above and below 5C were
11 and 30 kcal/mole, respectively. As
with Cyanidium, the decay below the
break point was composed of both a fast
and a slow component. The rates for the
slow component were used to plot Fig. 16.

Raison (1973), using spin labels with
isolated chloroplasts from chilling-sensi-
tive plants, has found that a phase transi-
tion occurs around 10°C in plants such
as bean and tomato; but no such transi-
tion was seen in chilling-resistant plants
such as pea or lettuce. We therefore
compared the temperature dependence of
the 520-nm change in the chilling-sensi-
tive plants bean and tomato with the
temperature dependence of this change
in chilling-resistant plants such as spinach
and lettuce.

Figure 17 compares the kinetics of the
520-nm change in tomato and lettuce
leaves at various temperatures. In the
tomato leaf the kinetics of absorbance
change during light exposure varied re-

A Ac,. 0.01

Tomafo 25 °C

Le+tuce

16 "C

7°C

4°C

t I

On Off

t i

t I

t i

T

AA5200.002

1

Fig. 17. Time courses of light-induced absorbance changes at 520 nm at different tempera-
tures in a tomato leaf grown at 25°C (upper part) and in a lettuce leaf grown at 15"C (lower
part). Repeated cycles of red light (described in Figure 14) and darkness were 5.0 and 7.5 sec,
respectively, for tomato and 3.5 and 9.0 sec, respectively, for lettuce.

230

CARNEGIE INSTITUTION

markably depending upon the tempera-
ture. However, the height of the initial
rise produced upon onset of actinic iUu-
mination was virtually unaffected by
changing the temperature. At 2°C and
also at 4~"X the increase of the second
slow phase of the 520-nm change dis-
appeared.

The lower part of Fig. 17 illustrates
absorbance changes at different tempera-
tures in a leaf of lettuce grown at 15°C.
By contrast to the tomato leaf, the time
courses were not so markedly affected by
changing the temperature. The second,
slow increase during light exposure could
still be seen at 4°C. We saw the same
tendency in the temperature dependence
of the time courses of the 520-nm change
in a lettuce leaf grown at 25 °C and in
market spinach leaves.

Figure 18 shows Arrhenius plots of
the dark decay of the 520-nm change in
tomato leaves grown at 15° or 25 °C. A
discontinuity point was found at about
lO^'C in the leaf grown at 15°C, and a
new slope began at about 12°C in the
leaf grown at 25°C. In a leaf grown at
35°C. the break appeared at 13°C (not
shown in the figure).

50
40

30
20

30

10

3.3

Tempera+ure, C
20 10

3.4 3.5

l/T X 10^ "K"'

3.6

Fig. 19. Arrhenius plots of the dark decay of
the absorbance change at 520 nm in leaves of
lettuce grown at 15° and 25 °C and in a spinach
leaf. Reciprocals of the half-decay time were
taken as representing the decay rates. Arrows
indicate the direction of temperature change.
Experimental conditions were as described for
lettuce in Figure 17.

I/T X 10^. °K"

Fig. 18. Arrhenius plots of the dark decay of
the absorbance change at 520 nm in leaves of
tomato grown at 15'" and 25'C. Reciprocals of
the half-decay time were taken as representing
the decay rates. Mea.surements were done by
decrea.sing the temperature. Experimental con-
ditions were a.s described for tomato in Figure
17.

A break was seen for a bean leaf at
15°C for both directions of temperature
change, although the two sets of lines
did not conform to each other.

Figure 19 shows the Arrhenius plots of
the dark decay of the 520-nm change in
leaves of lettuce grown at 15° and 25°C
and in a spinach leaf. There were no
breaks in these lines.

The light-induced absorbance change
at 483 nm in Cyanidium and the 520-nm
change in leaves of higher plants appear
to be due to an electrochromic shift
mainly of carotenoids (Reich et al., 1976;
Junge and Witt, 1968; Graber and Witt,
1976). After a fairly long period of illu-
mination (seconds), the spectral shift in-
duced in carotenoids is produced mainly
by ion gradients established by the influx
of H+ from the outer medium to the
inner space of the thylakoid (Jackson

DEPARTMENT OF PLANT BIOLOGY

231

and Crofts, 1969; Okada and Takamiya,
1970; Strichartz and Chance, 1972; N(3U-
mann and Jagendorf, 1964; Schuldiner
et al., 1972, 1973). The dark decay of
the carotenoid absorbance changes seems
to be correlated with the dissipation of
the ion gradient previously established
across the thylakoid membrane in the
light. In intact cells and leaves the dark
decays were rapid (about 100 msec) com-
pared to a 10-second decay rate seen in
isolated chloroplasts (Wakamatsu, et al.,
1974) . The more rapid decay in the intact
material is probably a reflection of a
close coupling of phosphorylation to the
dissipation of the ion gradient.

The breaks or the changes in the ap-
parent activation energies of the slow
decay component of the carotenoid change
seen in the Arrhenius plots in Cyanidium,
Tidestromia, and tomato suggest that
the coupling of electron transport to
phosphorylation is affected by a phase
transition in the membrane lipids. This
is supported by a recent finding of Ono
and Murata (unpublished) that a clear
break appears at the phase transition
temperature in the Arrhenius plot of pho-
tophosphorylation in thylakoid membrane
preparations of Anacystis nidulans.

Acknowledgments

We are grateful to Drs. Joseph A.
Berry and Olle Bjorkman and to Mr.
Jim Johnson, who provided the growth
cabinets and assisted in growing the let-
tuce and tomato plants. Dr. Joseph A.
Berry collected the Tidestromia plants
used in this study and Mr. Steven Graff
helped with the culture of Cyanidium.
Mr. Richard W. Hart built the cuvette
used in this study.

The Rate of Proton Gradient Decay

in Chloroplasts as an Indicator of

Membrane Parameters

Mordhoy Avron and David C. Fork

It has recently become clear that one
of the major functions of the thylakoid

membrane is to regulate proton passage
in such a way as to maximize the efficiency
of energy transduction by the photo-
chemical apparatus (Jagendorf, 1975;
Avron, 1977).

Previous reports from this laboratory
{Year Book 75, pp. 465-572; Murata et
al., 1975; Murata and Fork, 1975; Fork
and Murata, 1977) indicated that a
variety of techniques can be employed
to monitor changes in the physical phase
of the lipids in photosynthetic mem-
branes. In this study we investigated the
rate of dark decay of the pH gradient,
produced previously across the thylakoid
membrane by photosynthetic electron
transport, as a function of temperature
in chloroplasts isolated from both chilling-
sensitive and chilling-resistant plants.
We utilized the technique described by
Schuldiner et al. (1972) for following the
pH gradient across the thylakoid mem-
brane by observing the change in fluores-
cence of 9-aminoacridine in response to
the formation and decay of the gradient.

Chloroplasts were isolated by a stand-
ard technique (Avron, 1961) and were
suspended in a concentration of about
10 fxg chlorophyll/ml in a solution con-
taining Tricine, 15 mM, pH 8.0; NaCl,
20 mM; MgCU, 5 mM; phenazine metho-
sulfate, 10 fxM ; and 9-aminoacridine, 2
fiM. A monochromator and a narrow-
band interference filter provided a weak
blue light for exciting 9-aminoacridine
fluorescence. The fluorescence peaked at
400 nm and was monitored by a photo-
multiplier (EMI 9558B) protected by a
filter combination that peaked at 500 nm
and consisted of Corning filters CS 4-96
and CS 3-72, Calflex C, and a Wrat-
ten No. 64 filter. Strong red actinic light
was provided by filtering the light from
a tungsten-iodide lamp through a Calflex
C heat filter, 37 mm of water, and a
Schott glass filter RG 5 (3 mm thick).
A thermocouple in the reaction mixture
continuously monitored the temperature,
which was varied by passing cooled or
heated alcohol-water through a stainless
steel coil immersed in the reaction mix.

232

CARNEGIE INSTITUTION

The samples were heated and cooled at
rates of about TX and O.S'^C min, re-
spectively. The outputs of the photo-
multiplier and of the thermocouple were
fed directly into a Hewlett-Packard 2116
computer system, which calculated the
half-time of decay of the proton gradient;
its kinetics were monitored on a chart
recorder.

Figure 20 illustrates sample chart re-
cordings of the responses at different
temperatures and with a variety of chlo-
roplasts. The rate of decay of the proton
gradient was a strong function of tem-
perature — the rate for spinach chloro-
plastc? varied from a half-time of more

o

c

e

o

6)

Spinach

6.3 °C +, = 27;

Spinach

2.5 "C +, = 100:

Tidestromia

0.6 *C +, = 24 sec

1 2 3 4 5

Time, min

Fig. 20. The kinetics of light-induced proton
gradient formation and decay across chloroplast
thylakoid vesicles of spinach, bean, and Tide-
stromia oblongifolia at different temperatures.
The proton movements were followed at the
temperatures indicated as changes in the fluo-
re.scence of 9-aminoacridine excited by weak
blue light, as described in the text. The down-
ward and upward arrows mark the beginning
and end, respectively, of strong red actinic light
(^on for 52 sec) that excited the photosynthetic
reactions but was not detected by the fluores-
cence-measuring s>'stem. Gas phase, air.

1000

300

O
X

- 100

30

10 L-^

30

1 —

Temperature, C
20 10

-i 1 r

Spinach

3.3

3.4 3.5

l/T X 10^ °K'

3.6

Fig. 21. Arrhenius plot of the rate of proton-
gradient decay in spinach chloroplast fragments
previously illuminated by red actinic hght. Se-
quences of the light-induced decrease and the
following dark increase of 9-aminoacridine fluo-
rescence were measured as described in Fig, 20.
Light periods of 52 sec were followed by
sufficiently long dark periods to re-establish
the dark base level seen before illumination.
The sample at 16 °C was first cooled to 2.5° C
and then heated to about 30 °C.

than 100 seconds at temperatures around
0°C to fractions of a second at tempera-
tures around 35° C. Chloroplasts isolated
from Tidestromia oblongifolia or beans
(both chilling-sensitive plants) also ex-
hibited kinetics of proton efflux that were
strongly temperature dependent, but they
did not show the very slow decay char-
acteristically observed at low tempera-
tures with chloroplasts isolated from
spinach (chilling resistant).

An Arrhenius plot relating the recipro-
cal of the half-time of the decay of the
proton gradient to the reciprocal of the
absolute temperature for spinach chloro-
plasts is shown in Fig. 21. In agreement
with data previously obtained on the
520-nm shift and chlorophyll fluorescence

DEPARTMENT OF PLANT BIOLOGY

233

in spinach (Murata and Fork, 1975 and
this Year Book), no breaks were ob-
served between 20° and 2°C, and the
points in this region followed the same
line when temperature was decreased or
increased. The energy of activation in
this region is 11 kcal/mole. Above 20°C
the points deviated from linearity in a
manner similar to that previously de-
scribed for the temperature dependence
of atebrin fluorescence quenching (Kra-
ayenhof and Katan, 1971). However,
several lines of evidence indicate that
this break in linearity is not due to a
phase transition but rather to a high
temperature-dependent gradual inactiva-
tion of the chloroplasts: (1) the points
above 20° C do not fall on a straight line
but on a continuously upward-sloping
curve; (2) on decreasing the temperature
from points above 20°C, the "break" was
not observed, but instead the points fol-
lowed a straight line that represented

faster decay times than observed ini-
tially for a given temperature (see Fig.
22 for a similar situation in Tidestrowda
chloroplasts) ; (3) the temperature (20'"'
C in Fig. 21) at which the deviation from
linearity was observed was not constant,
but varied somewhat with different chlo-
roplast preparations (18-23''C) .

Arrhenius plots of the response in
Tidestromia chloroplasts are shown in
Fig. 22. Starting in the low-temperature
region we have repeatedly observed a
break in the plot around 5°C (see Fig.
22). This agrees with similar breaks seen
in the Arrhenius plot of the 520-nm ab-
sorption change and in the chlorophyll
fluorescence-versus-temperature curves
(Murata and Fork, this Year Book). This
break therefore also seems to indicate a
lipid phase transition at 5°C in this high-
temperature adapted plant. Above 5°C a
flat region was sometimes observed (see
Fig. 22, at right) extending for 5-8°C

1000

300

30

20

o
O

X

^ 100

X

C

o
"o

u

a

30

10

^^

Tempera+ure, C

20 10

7/

20

10

Tidestromia

J I I I I I I jii

it^

J I L

3.2 3.3 3.4 3.5 3.3 3.4 3.5

.3 o,.-l

^-IL

I I I I I L

3.4 3.5

3.6

3.7

l/T X 10 . °K"

Fig. 22. Arrhenius plots of the rates of proton-gradient decay from three different samples
of Tidestromia ohlongifolia chloroplast fragments. Measurements were as described in Fig. 21.
In each case the samples were initially at a temperature of about 7°C. One sample was
heated to about 36°C and then cooled again to 8°C (left). Another sample was heated to
30°C, then cooled (middle). Another sample was heated to 20°C and cooled again to — 1.6~C
(right). The Tidestromia, provided through courtesy of Dr. Olle Bjorkman, was grown at 32°C
day and 20^0 night temperatures.

234

CARNEGIE INSTITUTION

before a linear dependence was again
evident up to about 27-X. Within this
linear region, increasing or decreasing the
temperature produced identical results.
At temperatures above 27''C the char-
acteristic increasing slope, which we have
attributed to a temperature inactivation
(see the results with spinach chloroplasts,
above"*, was again evident.

In comparing the results obtained w^th
spinach chloroplasts to those observed
with Tidestromia chloroplasts, three fea-
tures can be emphasized: (1) Tidestromia
shows a transition around 5°C, which
seems to correlate with a membrane lipid
transition; no such transition is observed
in spinach. (2) The lowest temperature
at which irreversible inactivation is ini-
tially observed is around 20°C in spinach
and 27^C in Tidestromia. (3) In the low-
temperature region, spinach exhibits very
long half-times for proton efflux (> 90
sec), while the longest half-times ob-
served with Tidestromia were around 20
seconds.

An experiment on proton efflux was
also conducted with chloroplasts isolated
from leaves of bean, a chilling-sensitive
plant (Fig. 23). Similar characteristics
were observed — a break around 5°C, a
linear retraceable portion between 7-
17°C, and a temperature-dependent in-
activation like that observed with Tide-
stromia (not shown here in detail)
beyond 17°C. Here again, proton efflux
half-times at low temperatures were rapid
f< 20 sec).

It is clear from these results that the
rate of proton efflux through chloroplast
thylakoid membranes, as probed by the
fluorescence of 9-aminoacridine, is highly
temperature dependent and can be used
to monitor thylakoid membrane proper-
ties in a variety of plants. Species-specific
phase transitions and temperature-de-
pendent modifications of membrane
properties are parameters that can easily
be monitored with this technique.

The differences between chilling-sensi-
tive and chilling-resistant plants reported
in the previous article on the carotenoid

1000

Tempcra+ure, "C
10

O

X

300-

100-

3.4

l/T XIO"*. ^'K

Fig. 23. Arrhenius plot of the rate of proton-
gradient decay of bean chloroplast fragments.
Measurements were as described in Fig. 21. The
sample was heated from about 1.5 °C to about
20° C. The plants used for chloroplast extraction
were grown at 32 °C day and 20 °C night
temperatures.

change are also reflected by the behavior
of proton efflux as a function of tempera-
ture.

References

Allen, M. B., Arch. Mikrobiol. 32, 270-

277, 1959.
Amesz, J., Biochim. Biophys. Acta 79,

257-265, 1964.
Amesz, J., and L. N. M. Duysens,

Biochim. Biophys. Acta 64, 261-278,

1962.
Amesz, J., and J. W. M. Visser, Biochim.

Biophys. Acta 234, 62-69, 1971.
Anderson, J. M., Biochim. Biophys. Acta

416, 191-235, 1975.
Avron, M., Anal. Biochem. 2, 535-543,

1961.
Avron, M., Annu. Rev. Biochem. 46, 143-

155, 1977.

DEPARTMENT OF PLANT BIOLOGY

235

Brock, T. D., Nature 214, 882-885, 1967.

Castenholz, R. W., Bacteriol. Rev. 33,
476-504, 1969.

Cox, R. P., Biochim. Biophys. Acta 387,
588-598, 1975.

Dehlinger, P. J., P. C. Jost, and O. H.
Griffith, Proc. Nat. Acad. Sci. U.S.A.
7^2280-2284,1974.

Doemel, W. N., and T. D. Brock, J. Gen.
Microbiol. 67, 17-32, 1971.

Fork, D. C, and J. Amesz, Photochem.
Photobiol. 6, 913-918, 1967.

Fork, D. C, and N. Murata, in Photosyn-
thetic Organelles: Structure and Func-
tion, S. Miyachi, S. Katoh, Y. Fujita,
and K. Shibata, eds., Japanese Soc. of
Plant Physiologists, Tokyo, pp. 427-
436, 1977.

Graber, P., and H. T. Witt, Biochim.
Biophys. Acta 423, 141-163, 1976.

Inoue, H., and M. Nishimura, Plant Cell
Physiol. 12, 137-145, 1971.

Jackson, J. B., and A. R. Crofts, FEBS
Lett. 4, 185-189, 1969.

Jagendorf, A. T., in Bioenergetics of
Photosynthesis, Govindjee ed., Aca-
demic Press, New York, pp. 413-492,
1975.

Jost, P. C, 0. H. Griffith, R. A. Capaldi,
and G. Vanderkooi, Proc. Nat. Acad.
Sci. U.S.A. 70, 480-484:, 197d.

Junge, W., and H. T. Witt, Z. Naturforsch.
23b, 244-254, 1968.

Kempner, E. S., Science 142, 1318-1319,
1963.

Kleuser, D., and H. Biichen, Z Natur-
forsch. 24b, 1371-1374, 1969.

Kraayenhof, R., and M. B. Katan, in
Proc. 2nd International Congress on
Photosynthesis Research, G. Forti, et
al., eds.. Junk, The Hague, pp. 937-
949, 1971.

Meeks, J. C, and R. W. Castenholz,
Arch. Mikrobiol. 78, 25-41, 1971.

Murata, N., and D. C. Fork, Plant
Physiol. 56, 791-796, 1975.

Murata, N., and A. Takamiya, Plant Cell

Physiol. 10, 193-202, 1969.
Murata, N., J. H. Troughton, and D. C.

Fork, Plant Physiol. 56, 508-517, 1975.
Neumann, J., and A. T. Jagendorf, Arch.

Biochem. Biophys. 107, 109-119, 1964.
Nobel, P. 8., Planta 115, 369-372, 1974.
Ohnishi, S., and T. Itoh, Biochemistry

13, 881-887, 1974.
Okada, M., and A. Takamiya, Plant Cell

Physiol. 11, 713-721, 1970.
Oldfield, E., and D. Chapman, FEBS

Lett. 23, 285-297, 1972.
Ono, T., and N. Murata, Biochim.

Biophys. Acta 460, 220-229, 1977.
Raison, J. K., J. Bioenergetics 4, 285-

309, 1973.
Reich, R., R. Scheerer, K.-U. Sewe, and

H. T. Witt, Biochim. Biophys. Acta

449, 285-294, 1976.
Schuldiner, S., H. Rottenberg, and M.

Avron, Eur. J. Biochem. 25, 64-70,

1972.
Schuldiner, S., H. Rottenberg, and M.

Avron, Eur. J. Biochem. 39, 455-462,

1973.
Shneyour, A., J. K. Raison, and R. Smillie,

Biochim. Biophys. Acta 292, 152-161,

1973.
Strichartz, G. R., and B. Chance, Biochim.

Biophys. Acta 256, 71-84, 1972.
Traiible, H., Naturwissenschaften 58,

277-284, 1971.
Traiible, H., and P. Overath, Biochim.

Biophys. Acta 307, 491-512, 1973.
van Dijck, P. W. M., P. H. J. Th. Ver-

vergaert, A. J. Verkleij, L. L. M. van

Deenen, and J. de Gier, Biochim.

Biophys. Acta 406, 265-278, 1975.
Wakamatsu, K., and M. Nishimura, Plant

Cell Physiol. 15, 587-599, 1974.
Wakamatsu, K., N. Ikehara, and M.

Nishimura, Plant Cell Physiol. 15.

601-610, 1974.
Weikard, J., A. Miiller, and H. T. Witt,

Z. Naturforsch. 18b, 139-141, 1963.

236

CARNEGIE INSTITUTION

PROTON GRADIENTS AS POSSIBLE INTERMEDIARY

ENERGY TRANSDUCERS DURING ATP-DRIVEN REVERSE

ELECTRON FLOW IN CHLOROPLASTS

Mordhay Avron and Ulrich Schreiber

After light activation of the latent
adenosine triphosi^hatase. addition of
ATP in the dark drives reverse electron
flow in isolated chloroplasts, as evidenced
by reduction of the primary system II
electron acceptor, Q (Rienits et ai, 1974).
According to the chemiosmotic hypothe-
sis, proton gradients should be obligatory
energy-transducing intermediates in
ATP-induced reverse electron flow. In-
deed, it has been demonstrated that
transmembrane proton gradients are cre-
ated by the action of ATPase (Bakker-
Grunwald and Van Dam, 1973) and,
when artificially induced, drive reverse
electron flow by themselves (Shahak et
al, 1975). As an obligatoiy intermediate,
the buildup of a proton gradient is ex-
pected to be at least as fast kinetically
as the induction of reverse electron flow
under all experimental conditions.

We constructed an apparatus that
enabled us to monitor simultaneously the
ATP-driven buildup of ApH and reduc-
tion of Q under a variety of conditions.
Q-reduction was measured as the in-
crease of chlorophyll fluorescence yield,
with measuring light sufficiently low to
avoid light-induced Q-redox changes.
The same weak light excited 9-amino-
acridine fluorescence, the quenching of
which reflects the buildup of ApH

(Schuldiner et al, 1973). The results
are essentially in agreement with the
predictions of the chemiosmotic hypothe-
sis. Thus, slowing the ATP-induced
buildup of ApH by using a variety of
imcouplers and internal buffers and by
lowering the temperature brought about
a corresponding change in rate and extent
of the ATP-induced reduction of Q. Since
the observed ATP-driven reactions are
relatively slow (half-time of a few sec-
onds), they may permit complete equili-
bration between other postulated high-
energy intermediates and the observed
proton gradients. This material is pub-
lished in detail elsewhere (Avron and
Schreiber, 1977).

References

Avron, M., and U. Schreiber, FEES Lett.

77, 1-6, 1977.
Bakker-Grunwald, T., and K. Van Dam,

Biochim. Biophys. Acta 292, 808-814,

1973.
Rienits, K. G., H. Hardt, and M. Avron,

Eur. J. Biochem. 43, 291-298, 1974.
Schuldiner, S., H. Rottenberg, and M.

Avron, Eur. J. Biochem. 39, 455-462,

1973.
Shahak, Y., H. Hardt, and M. Avron,

FEBS Lett. 54, 151-154, 1975.

ATP-INDUCED CHLOROPHYLL LUMINESCENCE IN
ISOLATED SPINACH CHLOROPLASTS

Ulrich Schreiber and Mordhay Avron

After light-activation of the latent tion of Q by reverse electron flow (Avron

ATPa>f'. arlflition of ATP in the dark to and Schreiber, 1977; see also Avron and

isolated chloroplasts leads to formation Schreiber, this Year Book). Both the

of a transthylakoidal ApH and reduc- ApH and the reduction of Q constitute

DEPARTMENT OF PLANT BIOLOGY

237

conditions favorable for the stimulation
of post-illumination luminescence. This
luminescence is generally considered to
result from the recombination of positive
(Z+) and negative {Q") charges at the
photosystem II reaction centers (Lavorel,
1975). Although stimulated luminescence
has been reported when a ApH and Q-
reduction was artificially induced by an
acid-base transition (Shahak et al., 1977) ,
attempts to observe ATP-induced lumi-
nescence under the particular conditions
required for ATPase activation have
been unsuccessful.

In the present study we developed a
procedure by which ATP-induced lumi-
nescence can be readily measured. This
procedure is illustrated in Fig. 24. The
chloroplasts were first illuminated with
strong, heat-filtered white light for three

minutes to activate the latent ATPa.se.
Shortly after th(; activating light was
turned off, the decay of post-illumination
luminescence was monitored for 90 sec-
onds, by which time it approached zero.
Then two saturating flashes were given,
and the flash-induced luminescence was
recorded for 90 seconds. Another flash
pair was then given, and ATP was in-
jected before the flash-induced lumines-
cence was completely decayed. ATP
caused about a tenfold stimulation of
luminescence. A third flash pair, given
90 seconds after the second in the pres-
ence of an active ATPase, induces again
about ten times more luminescence than
could a flash pair in the absence of ATP
(see dotted line).

The requirements for observing a sig-
nificant ATP-induced luminescence are

>^

c

0)
u

c

u
(/)
0)

c

E

_2

>

6.0 6.5

Time, min

7.5

8.0

Fig. 24. ATP-induced stimulation of luminescence in isolated spinach chloroplasts. The light-
off arrow indicates the end of the 3-min light activation period. At the zig-zag arrows, pairs
of brief (peak intensity after 2 ^sec) saturating flashes (Stroboslave) were given. The broken
lines represent the decay of flash-induced himinescence in an identical sequence without ATP
addition. Monitoring of postillumination luminescence was initiated 5 sec after preillumination.
ATP addition was by injection of 3 /ul 0.1 M solution into 500 ^1 of stirred reaction mixture.
The measuring apparatus and measuring conditions were as described previously (Avron and
Schreiber, 1977), except for a much lower PMS concentration (here 5 X 10"^ M).

23S

CARNEGIE INSTITUTION

listed in Table 2. In contrast to the ATP-
indiiood reduction of Q (Rienits et al.,
1974). PMS does not seem to be obliga-
tory in this system. Actually the same
high concentrations of PMS (> 1 /xM) ,
which are optimal for ATP-induced re-
verse electron flow (Rienits et a/., 1974),
accelerate the decay of flash-induced
luminescence to a point that ATP-induced
luminescence cannot be observed.

An important observation was that
unless at least one preilluminating flash
were given prior to ATP addition, almost
no luminescence was stimulated by ATP.
This result is in agreement with the
postulated mechanism of luminescence
by recombination of the positive and
negative charges separated by the photo-
act at system II reaction centers (Lavorel,
19751. Without a preilluminating flash,
luminescence is limited by the availability
of positive charges, since the Z-pool is
relatively small compared to the Q-PQ
pool. ATP does not stimulate lumines-
cence in this situation, because the ATP
effect is on the Q-PQ side. A saturating
flash creates an equal number of positive
and negative charges. While the negative
charges drain rapidly into the secondary

TABLE 2. Requirements for ATP-induced

Luminescence*

Extent of ATP-induced

Luminescence

Conditions

(% of control)

No light activation

1

DTT omitted

15

PMS omitted

94

MgClo omitted

MgCL omitted during

activation only

49

Xo flashes

8

One flash

32

* The values given represent the relative
height of the luminescence peak obtained upon
ATP injection. The complete reaction mixture
and control procedure are described in the text.
When MgClz was omitted during activation
only, it was added 15 sec after the activating
light was turned off. No flashes: ATP was
added when the postactivation luminescence
reached the same level as in the control.

acceptor pool, the positive charges have
a relatively long half-life, on the order
of 30 seconds (Joliot et al, 1971). In
this situation an ATP-induced reverse
electron flow from the secondary acceptor
pool back to Q can lead to marked stimu-
lation of luminescence.

>s

0)

u

c
cy

m

c

e

_2
>

cr

Fig. 25. Dependence of ATP-stimulated luminescence on the time elapsed between a pre-
illuminating flash pair and ATP-injection. Conditions and preillumination schedule were as in
Fig. 24.

DEPARTMENT OF PLANT BIOLOGY

239

An important parameter for the yield
of ATP-induced luminescence is the time
elapsed between the flashes and ATP
addition (Fig. 25). The phenomenon is
the more pronounced the earlier ATP is
added, but a significant stimulation is
observed even one minute after the
flashes, when the flash-induced lumines-
cence itself is almost completely decayed.

If the phenomenon of ATP-induced
luminescence is indeed due to the ATP-
synthase system working in reverse, it
should be sensitive to uncouplers and
ATPase inhibitors. These predictions are
confirmed by the results shown in Fig. 26.
Gramicidin, which is a potent uncoupler,
and tentoxin, an ATPase inhibitor
(Steele et al., 1976), each completely
abolished the effect and left only the
flash-induced luminescence.

These experiments demonstrate that
after proper activation the ATPase sys-
tem can affect the electron transport
system all the way to the photosystem II
reaction center. Thus they extend and

complement the previous demonstrations
of an ATP-induced proton gradient and
ATP-induced reduction of Q (Avron and
Schreiber, 1977, and this Year Book).

References

Avron, M., and U. Schreiber, FEES Lett.
77, 1-6, 1977.

Joliot, P., A. Joliot, B. Bouges, and G.
Barbieri, Photochem. and Photobiol.
14, 287-305, 1971.

Lavorel, J., in Bioenergetics of Photo-
synthesis, pp. 319-371, Govindjee, ed.,
Academic Press, New York, 1975.

Rienits, K. G., H. Hardt, and M. Avron,
Eur. J. Biochem. 43, 291-298, 1974.

Shahak, Y., Y. Siderer, and M. Avron, in
Photosynthetic Organelles, Special Is-
sue of Plant Cell Physiol., pp. 115-127,
Center for Academic Publications,
Japan, 1977.

Steele, J. A., T. F. Uchytil, R. D. Durbin,
P. Bhatnagar, and D. H. Rich, Proc.
Nat. Acad. Sci. U.S.A. 73, 2245-2248,
1976.

>N

(X)

o

c

O

c
£

>

"T r

Gramicidin

ATP

Tenfoxin

ft t

ATP

2 3 4

Time, mm

Fig. 26. Inhibition of ATP-stimulated luminescence by gramicidin and tentoxin. One pre-
illuminating flash pair was given before ATP-addition, 2 min after light-activation. The
inhibitors were added 15 sec after activation. Concentrations: gramicidin, 5 X 10"^ M ; tentoxin,
10"^ M. The tentoxin was a gift from Dr. J. A. Steele, University of Wisconsin.

240

CARNEGIE INSTITUTION

STUDIES ON DNA SEQUENCE ORGANIZATION

Michael G. Murray. Richard S. Preislcr, and William F . Thompson

Last year we reported preliminary
studies which established that the pea
genome is organized in a pattern gen-
erally similar to that of many animal
genomes [Year Book 75, p. 356). Single-
copy sequences were shown to be less
than 3.000 nucleotides in length, with
both short (200-400-nucleotide) and long
(> 1500-nucleotidel repetitive sequences
present. The longer repeats reassociate to
produce duplexes with high thermal sta-
bility, indicative of precise base pairing,
while the short repeats form duplexes
with greater mismatch, indicating that
there has been more base sequence di-
vergence within families of short repeti-
tive sequences.

More detailed studies of both pea and
mung bean DNA are now in progress.
Experiments have been designed to ex-
amine the interspersion of single-copy
sequences with repetitive DNA, applying
techniques developed by Davidson et at.
(19731 and Graham et al. (1974). These
experiments rely on the fact that a duplex
region in an otherwise single-stranded
DNA molecule wall cause the entire
molecule to bind to hydroxylapatite.
Thus if DNA is reassociated so that only
repetitive sequences can react, the bound
fraction will include any unreacted single-
copy DNA present in the same fragment
with a repeated sequence. If the frag-
ments employed are shorter than the
single-copy sequences, most of the single-
copy DNA will be found on fragments
containing no repetitive elements, and
will therefore not bind to hydroxylapatite.
However, as the fragment length is in-
creased, more and more single-copy DNA
will become bound, since the fraction of
fragments containing at least one repeti-
tive element will increase with increasing
fragment length. Binding will increase

until the fragments exceed the length of
single-copy sequences. Further increases
in length will result in fragments con-
taining two repetitive regions; however,
this will not increase the extent of hy-
droxylapatite binding since a single du-
plex region is sufficient to cause binding.
If many single-copy sequences are
contained in a class with a relatively
narrow distribution of length, there will

1.0

as

0.4

"1 r

n — I 1 — r

Pec

J I I L

J I I I L

500

1000

1500

Fragment length, nucleotides

Fig. 27. The fraction of pea DNA fragments
containing repetitive-sequence elements, as a
function of fragment length. Labeled DNA
fragments of the indicated average lengths were
reassociated with a 2500-fold excess of 400-
nucleotide fragments to Cot 50. The fraction
bound to hydroxylapatite (R), was corrected
for zero time binding as described in the text
and plotted as a function of fragment length.
Fragment lengths were determined after incuba-
tion. Values shown are averages of three
repHcations and vertical lines denote standard
deviations.

DEPARTMENT OF PLANT BIOLOGY

24

be a sharp break in the slope of the
binding-versus-length curve at a point
corresponding to the average length of
that class (Davidson et al., 1973;
Graham et al., 1974). In most higher
organisms studied, a majority of the
single-copy regions are shorter than a few
thousand nucleotides, a condition reflect-
ing the interspersion of repeated se-
quences in a short-period pattern; a
smaller number are much longer, reflect-
ing long-period interspersion of repeats
(Davidson et al., 1975).

Figures 27 and 28 show the results of
one series of experiments on pea and
mung bean DNA. DNA was labeled by
iodination with Na^^^I as described

1.0 -

0.9

_ 0.8
q:

-| 1 — I r

-] — I 1 — I — I 1 I T r

Mung bean

0.41 — ^ — L.

J I I L

J L

500 1000 1500

Fragment length, nucleotides

Fig. 28. The fraction of mung bean DNA
fragments containing repetitive-sequence ele-
ments as a function of fragment length. Labeled
fragments of the indicated lengths were incu-
bated with a 2500-fold excess of 400-nucleotide
fragments to Cot 100 and then treated as de-
scribed in the legend to Fig. 27. Values shown
are averages of three replications and the
vertical lines denote standard deviations.

previously (Year Hook 76, p. 363 j, ex-
cept that the reaction was carried out
for 1-2 min at 98'' tcj prevent reassoci-
ation or intrastrand duplex formation.
The labeled DNA was then fractionated
on a preparative alkaline sucrose gradi-
ent to obtain tracer preparations of dif-
ferent lengths, and the different tracers
were allowed to react separately with an
excess of unlabeled fragments 400 nucleo-
tides long. Binding was measured at C<,t
50 for pea DNA and Cut 100 for mung
bean. Figures 29 and 30 show that almost
all repetitive sequences have reacted by
the time these C^t values are reached,
while little or no single-copy DNA has
reassociated. Fragment lengths were
measured by electrophoresis in alkaline
agarose gels (Murray and Thompson,
this Year Book) after reassociation in
order to take into account any strand
scission that may have occurred during
the incubation at elevated temperature.
Total binding (B) was corrected to elimi-
nate the effect of ^'zero time" binding
elements (foldback or inverted repeat
sequences) by measuring the fraction
(Z) of each tracer bound after incuba-
tion to Cot 10~^. The fraction bound
because of bimolecular reassociation be-
tween repeated sequences is then:

R^lOOiB -Z)/{1 -Z)

This correction involves the assumption
that sequences adjacent to '^zero time"
binding elements behave like other se-
quences in the genome.

It is readily apparent from the figures
that the data could be improved by
including several longer and some shorter
tracer lengths. Such work is currently in
progress. Several important conclusions
can already be drawn. By extrapolation
to zero fragment length, the actual frac-
tion of genome consisting of repetitive
sequences can be estimated as about 0.63
for pea and 0.44 for mung bean. In the
case of pea DNA, this value is consistent
with results from experiments using SI
single-strand specific nuclease {Year

242

CARNEGIE INSTITUTION

Log equivalent Cq\

Fig. 29. (A) Reassociation kinetics of short pea DNA fragments. Samples at various concen-
trations in either 0.12 or 0.40 M sodium phosphate buffer, pH 6.8, were heat-denatured and
incubated at 60' or 66°, respectively, for various times prior to fractionation on hydroxylapa-
tite in 0.12 M sodium phosphate at 60°. Cot values for samples incubated in 0.40 M sodium
phosphate have been corrected to the equivalent Cot values in 0.12 M sodium phosphate at
60". Data are from independent experiments using unlabeled DNA preparations with average
lengths of 360 or about 400 nucleotides. Tracer quantities of ^H-DNA were prepared from sheared
single-stranded total DNA using E. coli DNA polymerase I and ohgonucleotide primers, and
the reaction product was treated with SI nuclease prior to mixing with unlabeled driver frag-
ments (Murray, Belford, and Thompson, this Year Book). The upper curve is a least-squares
fit to the pooled data, using the three theoretical second-order components shown, plus a 9%
"ver>' fast" fraction reacting prior to Cot 0.01. The rate constant for single-copy sequences was
fixed at 2.4 X 10"* 1/mol-sec, the value predicted for a haploid nuclear DNA content of 4.5 pg.
Solid circles indicate unlabeled DNA; open circles, ^H polymerase copy DNA. (B) Reassocia-
tion kinetics of long pea DNA fragments. Tracer fragments with an average length of 1200
nucleotides were mixed with a 2500-fold excess of unlabeled 400-nucleotide fragments and
reassociated as described for Fig. 29 A. At each Cot value, the fraction of tracer fragments con-
taining renatured regions was measured by hydroxylapatite fractionation. The upper curve is a
least-s-juares fit for the two repetitive sequence components shown, plus a 15% 'Very fast"
component reacting prior to Cot 0.01. The tracer used in this experiment was obtained from
the .«:ame ^^I-DNA preparation as the fragments of other lengths used in the experiment of
Fig. 27. Its length was estimated by electrophoresis in an alkaline agarose gel after incubation
in 0.40 M sodium phosphate for 6 hr (equivalent Cot =. 50).

Book 75, pp. 356-362), and also with
hypochromicity measurements on re-
associated DNA (not shown).

The data also show that most of the
single-copy sequences in pea DNA are

present in a short-period interspersion
pattern (evident from the steep slope in
the initial phase of the binding curve in
Fig. 27) ; the mean length for this class
falls between 600 and 900 nucleotides.

DEPARTMENT OF PLANT BIOLOGY

243

0.2
0.4 h
0.6

T5

0.8

o

_Q

C

1.0

o

u

o

"1-

0.2 -

0.4 -

0.6 -

0.8 -

B

-2

I 2

Log equivalent Cof

Fig. 30. (A) Reassociation kinetics of short mung bean DNA fragments. Mung bean DNA
sheared to an average length of about 400 nucleotides was reassociated and fractionated as
described in Fig. 29A. The upper curve represents a least-squares fit to the data using the two
theoretical second-order components shown, plus a 15% "very fast" fraction. (B) Reassociation
of ^^^I mung bean DNA fragments with an average length of 1200 nucleotides in the presence
of excess unlabeled 400-nucleotide fragments. Reassociation and fractionation were carried out
as described for Fig. 29B. The upper curve is a least-squares fit with the two theoretical second-
order components shown, plus a 24% "very fast" fraction. The rate constant for the single-copy
fraction of the tracer was fixed at 9 X 10"* 1/mol-sec (three times that observed for the
400-nucleotide fragments), which is the value expected for 1200-nucleotide fragments.

This value compares with estimates rang-
ing from ^800 to 1500 nucleotides for
animal genomes (Davidson et al., 1975).
Walbot and Dure (1975) estimated the
mean length of short-period single-copy
DNA in cotton to be about 1800 nucleo-
tides, while Zimmerman and Goldberg
(1977) found a value of about 1400
nucleotides for tobacco DNA. Some-
what shorter sequences, averaging around
1000 nucleotides, were found in wheat
DNA by Flavell and Smith (1976) , while
in rye DNA the major class of inter-
spersed single-copy elements may aver-
age only about 400 nucleotides in length
(Smith and Flavell, 1977). Since our

estimates of tracer fragment lengths were
made after reassociation, we feel that
ours more accurately reflect the situation
at the time of the binding assay than do
the measurements of initial tracer lengths
used in many of the other studies. Control
experiments show that strand scission
reduces the average length of 1200-1400-
nucleotide tracers by about 20-25%
under our conditions.

By extrapolating the data for the
longer fragment lengths to the y-axis.
it is possible to derive an estimate of
16% (100 — 84) for the fraction of the
pea genome composed of long-sequence
elements unreacted at Cot 50. About 21%

244

CARNEGIE INSTITUTION

of the genome (84 — 63) would then
be composed of sequences in the short-
period interspersion pattern. However,
these calcuhitions fail to take into ac-
count either any unreactable hibel in the
tracer preparations or the presence of re-
petitive sequences whose reassociation is
not yet complete at Cot 50. The reassoci-
ation kinetics of 1200-nucleotide tracer
DXA. presented in Fig. 29 (lower) , reveal
that about o^c of the DNA fragments of
this length contain repetitive elements
that, although they have not yet bound
by Cot 50. do however bind at Cot values
too low for reassociation of single-copy
DXA. In addition, about 3% of the label
appears to be unreactable. If we assume
the interspersion of unreacted repetitive
sequences to be the same as that of the
total repetitive sequence population, the
maximum binding we could expect at Cot
50 would be 92%. Normalizing the data
in this fashion lowers the estimate of the
fraction of the genome in the long-period
interspersion pattern to about 10 percent.

These calculations are intended only
to illustrate the uncertainty in estimat-
ing the fraction of long-period single-
copy DXA in highly repetitive and
extensively interspersed genomes. In fact,
the more sensitive kinetic analysis shown
in Fig. 29, bottom panel, failed to detect
any .^ingle-copy sequences longer than
12(X) nucleotides. At least some of the
increased binding observed in Fig. 27 at
lengths above 700 nucleotides or so may
be attributed to heterogeneity of the
single-copy sequence lengths and of the
tracer fragment lengths. It is thus possi-
ble that single-copy sequences with a
mean length greater than 600-900 nucleo-
tides do not exist as a discrete class in
the pea genome.

Experiments such as that summarized
in Fig. 27 mea.sure the binding that re-
sults from complete reassociation of all
repetitive sequences, and thus provide
no information on the arrangement of
repeater] sequences in other frequency
classes. In order U) study interspersion
of different frequency classes, we have

therefore constructed Cot curves describ-
ing the reassociation kinetics of long
(1200-nucleotide) tracer DNA fragments
in the presence of an excess of short frag-
ments. By comparing the fractions of
long and short fragments reassociating
with the kinetics of a given repetitive
component, we can estimate the extent to
which sequences belonging to one fre-
quency class are interspersed among
more slowly reacting sequences in the
genome. Results of two such experiments
with pea and mung bean DNA are pre-
sented in Figs. 29 and 30. Theoretical
second-order kinetic components were
fitted to the data by means of a least-
squares computer program, and the re-
sults of this analysis are summarized in
Table 3.

In the case of pea DNA, both the "very
fast" and ''fast" components include a
larger fraction of the long tracer frag-
ments and thus are seen to be inter-
spersed among more slowly reacting se-
quences. Sequences adjacent to elements
of the very fast fraction may include
fast and ''slow" middle-repetitive as well
as single-copy sequences, but do not con-
stitute a large fraction of the total DNA.
By contrast, the fast fraction clearly
dominates the long tracer reaction, and
sequences in this class must therefore be
interspersed throughout much of the
genome. About 70% of the long tracer
molecules contain at least one fast repeat
element, as opposed to about 25% in the
driver. Almost half of the pea genome
must therefore be composed of slow and
single-copy sequences in close proximity
to fast sequences. Extensive interspersion
of slow and fast repeats would explain
the fact that the slow component ac-
counts for only about half as many long
tracer fragments as it does driver frag-
ments.

It seems likely that the remaining slow
repeats are interspersed among single-
copy sequences. Long tracer fragments
containing only slow sequences would be
expected to reassociate faster than the
slow repeats in the short driver DNA by

DEPARTMENT OF PLANT BIOLOGY

245

TABLE 3. Comparison of Theoretical Components Fitted to Reassociation Kinetics Data
from Figs. 3 and 4 for Long-Tracer and Short-Driver DNA Fragments

Driver

Tra

cor

Fraction

Rate

Fraction

Rate

Genome

Components

(Qa)

(K.)

(Q,)

(Kr)

Q, - Q^i

Qt/Qi

Pea

Very fast*
Middle repetitive

0.09

0.15

...

+0.06

1.7

(a) fast

0.24

3.3

0.71

4.9

+0.47

3.0

(b) slow

0.36

0.05

0.16

0.05

-0.20

0.4

Single-copy

0.31

0.00024t

-0.31

Mung Bean

Very fast*

0.15

0.24

+0.09

1.6

Middle repetitive

0.34

0.09

0.51

0.08

+0.17

1.5

Single copy

0.51

0.0003

0.25

0.0009t

-0.26

0.49

* The "very fast" fraction includes "foldback" or inverted repeat sequences as well as any
sequences reassociating prior to Cot c^ 10"^.

t Single-copy rates fixed at predicted values during computer analysis.

a factor of at least 3 (the tracer/driver
length ratio; Davidson et al., 1973). We
detect no significant difference in the rate
of the slow component in our data, and
therefore we assume that most fragments
in the slow tracer component probably
consist of short repetitive elements ad-
jacent to single-copy DNA. However,
given the uncertainties inherent in fitting
theoretical components to complex re-
association data, this conclusion must be
regarded as tentative. Additional experi-
ments are clearly required before a firm
conclusion will be possible.

Although mung beans also belong to
the family Legwminosae, the data pre-
sented in Figs. 28 and 30 indicate that
the pattern of sequence organization in
mung bean DNA is strikingly different
from that of pea DNA. Single-copy
sequences account for a greater propor-
tion of the total DNA in mung beans
(about 59% versus 37% in pea DNA)
and most of the mung bean single-copy
sequences are significantly longer. The
break in the binding curve for mung
bean DNA fragments (Fig. 28) is less
distinct than that for pea DNA, but
would appear to be in the vicinity of
1000 nucleotides. By estimating the frac-
tions of short- and long-period single-

copy sequences by extrapolation as de-
scribed for pea DNA, we obtain values
of 14% and 42% of the genome respec-
tively. Using the data of Fig. 30 to
estimate about 4% unreacted repetitive
sequence at Cot 100 and 6% unreactable
label, these figures can be normalized to
19%) and 47%^. Thus, roughly 70%
(47/66) of the mung bean genome con-
sists of long-period single-copy sequences.
This contrasts markedly with pea DNA,
in which such long single-copy sequences
cannot even be conclusively demonstrated
to exist.

The reassociation kinetics of long
(1200-NT) mung bean tracer with short
driver can best be fit with two second-
order components, one repetitive and one
non-repetitive (Fig. 30, bottom panel).
About 25% of the 1200-nucleotide tracer
reassociates according to single-copy
kinetics, indicating that about half the
single-copy sequences are longer than
1200 nucleotides in the mung bean ge-
nome. Most of the remaining single-copy
DNA must be interspersed with middle-
repetitive sequences, as indicated by the
increase in the fraction of fragments re-
associating with this component in the
long tracer. Some of the single-copy
and/or middle-repetitive sequences also

246

CARNEGIE INSTITUTION

appear to be interspersed with sequences ment length of 1200 nucleotides. In mung
in the "very fast" fraction, although (as beans, the major class of repetitive se-

in pea DXA) these sequences do not
constitute a major fraction of the genome.
In summarizing our results on the pea
and mung liean genomes we are impressed
bv both similarities and differences. Both

quences is similar to the slow repeated
component in peas and is interspersed
primarily with single-copy DNA.

We believe that further studies of se-
quence organization and transcription in

genomes are organized according to the such different genomes may help to de-

general pattern which characterizes the
DXA of most animal species examined
(Davidson, et al., 1975). In both species
a significant fraction of the genome is
organized in a short-period interspersion
pattern, in which single-copy sequences
average 700-1000 nucleotides in length.
Much of the interspersion observed with

termine which aspects of sequence or-
ganization have functional significance.

References

Davidson, E. H., B. R. Hough, C. S.
Amenson, and R. J. Britten, J. Mol.
Biol 77, 1-23, 1973.

1200-nucleotide fragments in both ge- Davidson, E. H., G. A. Galau, R. C.
nomes is attributable to middle-repetitive Angerer, and R. J. Britten, Chromo-
sequences. However, the mung bean ge- soma 51, 253-259, 1975.

nome contains a larger fraction of single- Flavell, R. B., and D. B. Smith, Heredity
copy DXA than does the pea genome, 37, 231-252, 1976.
and a much larger fraction of the mung Graham, D. E., B. R. Neufeld, E. H.
bean single-copy sequences are arranged Davidson, and R. J. Britten, Cell 1,
in a long-period interspersion pattern. 127-137, 1974.

Pea DXA contains a large component Smith, D. B., and R. B. Flavell, Biochim.
of rapidly reassociating repeated se- Biophys. Acta J^7I^, 82-97, 1977.
quences in addition to a more slowly Walbot, V., and L. S. Dure, J. Mol. Biol.
reassociating component. These two re- 101, 503-536, 1976.
peated sequence classes are significantly Zimmerman, J. L., and R. B. Goldberg,
interspersed with one another at a frag- Chromosoma 59, 227-252, 1977.

5IXGLE-COPY DNA SEQUENCE COMPARISONS IN Atriplex
Heather S. Belford and William F. Thompson

In the past year we have continued
the single-copy DX^A sequence compari-
son studies in the genus Atriplex. The
number of species included in the survey
was larger than previously, and some
major technical modifications were made.
Our objective has been to investigate the
apparent disagreement between the clas-
sical phylogenetic interpretation (Hall
and Clements, 1923) and the interpreta-
tion derived from recent genetic (Nobs,
Year Book 75, pp. 421-423, 1976) and
molecular (Belford and Thompson, Year
Book 75, pp. 362-367) hybridizations.

The classical scheme is a double-branched
tree in which the C4 photosynthetic spe-
cies of both sub-generic branches evolved
from C3 ancestors. The current data
suggest instead that the two main species
groups are quite distinct in their evolu-
tionary heritage.

The species included in our present
work are: A. hortensis (C3), A. triangu-
laris (C.i), A. sabulosa (C4), andvl. rosea
(C4) from the subgenus Euatriplex; and
A. phyllostegia (C3) and A. serenana
(C4) from the subgenus Obione. A.
truncata (C4) (Obione) DNA has been

DEPARTMENT OP PLANT BIOLOGY

247

prepared for inclusion in future experi-
ments. These species have been chosen to
represent major species groups in the
classical scheme. In addition, A. fruti-
culosa, a species of Obione which forms
fertile hybrids with A. serenana (Nobs,
Year Book 75, pp. 421-423), has been
included to provide a ''control" combina-
tion of species known to be closely re-
lated genetically. A similar combination
will be obtained in future experiments
between A. rosea and A. sabulosa.

DNA has been extracted from each
species and sheared to a length of about
400 base pairs in a Vir-Tis "60" homog-
enizer (Britten et al, 1974). DNA highly
enriched for single-copy sequences was
prepared by heat-denaturing and re-
associating DNA (in 0.12 M sodium
phosphate buffer at 60°C or 0.6 M
sodium phosphate buffer at 66°C) to an
equivalent Cot of 350. After reassociation
the mixtures were, diluted to 0.12 M
sodium phosphate and chromatographed
on hydroxylapatite at 60°C. Approxi-
mately 25% of the total OD260 was re-
covered in the unbound fraction as un-
reassociated, primarily single-copy se-
quence DNA. The single-copy fraction
was purified over AG-50WX8 cation ex-
change resin (Bio Rad) and chelex-100
and then used to prepare radioactively
labeled sequences for molecular hybridi-
zations.

In the current studies, the in vitro
iodination procedure previously used was
replaced by an oligonucleotide-primed
E. coli polymerase I reaction that allows
incorporation of ^^C- or ^H-labeled de-
oxyribonucleotide triphosphates. This
technique permits double-labeled experi-
ments in which both homologous and
heterologous tracer reactions may be fol-
lowed in the same mixture. The homolo-
gous reaction serves as an internal con-
trol in each experiment, thus eliminating
some of the variability possible between
separate experiments. Radioactively la-
beled E. coll polymerase I copies were
made from Atriplex single-copy sequences
according to the standard protocol de-

tailed in this Year Book, (Murray, Bel-
ford and Thompson). Each reaction gave
an average of 60-70% copy with a frag-
ment length of 250-300 base pairs as:
determined by alkaline agarose gel elec-
trophoresis. The DNA copy was purified
from the enzyme mixture by hydrox-
ylapatite chromatography in 8 M urea
(Britten et ai, 1974). About 25% of the
labeled product renatures extremely
quickly even at very low concentrations,
indicating that it contains intrastrand
complementary sequences. Fragments
containing these intrastrand comple-
mentary or ''foldback" sequences can be
removed (''stripped") from the purified
copy DNA by denaturation followed by
rapid cooling to 60°C and immediate
hydroxylapatite fractionation. Labeled
copies were prepared from single-copy
DNA fractions for each of the eight
Atriplex species, purified, and stripped
of foldback to yield clean-copy DNA
with specific activities of approximately
6.6 X 10^ cpm/fxg.

In previous Atriplex hybridization ex-
periments the species from which the
radioactive tracer DNA was obtained
was held constant while the driver DNA
species was varied. To continue this
strategy in the current expanded study
would involve the extraction from each
species of large quantities of DNA, which
was impractical in several cases. There-
fore, we have modified our experimental
design to vary the tracer DNA species
while holding the driver species constant.

To date, we have hybridized single-
copy tracers from A. fruticidosa, A. phyl-
lostegia, A. rosea, A. sabulosa, A. tri-
angularis and .4. hortensis with an excess
of unlabeled total DNA from A. serenana.
Each interspecies cross has been ana-
lyzed in terms of complementarity within
homologous sequences and extent of cross
reactivity. Mismatch estimates were
derived from the thermal denaturation
("melting") profiles of the hybrid mole-
cules to provide a measure of comple-
mentarity. Any degree of mismatch in a
duplex molecule lowers the Tm (the tern-

~0

c

D

o

_Q

C

o

u
D

1
- oo- -

1111

I
A. serenana

-

•

-

a2

-

• .N^ OO *^0 O

"

-

>v " O* ^0

-

0.4

-

\ . ° "xO

-

-

^Vv "^^ ^o o

-

0.6

"

^-i->l^. ^^^^°^

X 8

*

^

"X..,,^ \

«i

- •

- . -^S^ s

X

d^DOQ O

0.8

^, o^'^-og O O O

1.0

1

1 1 "" - 1- 1

^^^^V-OJ,^.

A. horfensis

-

" ■ — *w^ ^

0.2

~

^

~

X
X
X

^^^^.^^

0.4

■-

— ^— _

-

-

X

•

-

0.6

*~

\

"■

— ' ....____^ ^

-

N

"

0.8

r ■

-

1.0

1

1 1 1 1

A. fruficulosa

-

'* ~"^~^ — — — ^

** •• >

"

-^^-~v^

0.2

h-

"

'

■

0.4

-

^S^ X X.

N. X >

\ X

-

06

-

\ X
\ X

X X

\^ X

N!

~

^N^ ^^""•"V— -«—

0.8

-

1 n

1

^1 ■ ~~ ~-r--_ , ,

I 2 3

Log equivalenf Co+

Fig. 31. Rea.ssociation kinetics for total cell DNA from A. serenana and single-copy tracers
from A. serenana, A. horlensin, and A. jruticulosa. Top panel: Driver (•) and tracer (O)
kinetics for A. serenana self-reactions. Data were pooled from four independent experiments.
Curves through the data points represent least-squares fits to the data using two theoretical
second-order components for the tracer data and three for the driver. The theoretical compo-
nents from the tracer fit are indicated by the light dashed lines. Middle and bottom panels:
Reassociation of tracers from A. hortensis and A. jruticulosa in the presence of excess A.
serenana driver DNA. Lines through the data points represent least squares fits for two com-
ponenta, u.sing the equation for retarded kinetics described in the text. The theoretical com-
ponents of each fit are represented by solid lines in the lower part of each panel. Dashed lines
representing the reaction of A. serenana tracer have been included to facilitate comparison.

DEPARTMENT OF PLANT BIOLOGY

249

perature at which half of the duplex
molecules are separated into single
strands) below that for native DNA. The
/\Tm between homologous and heterolo-
gous hybrid molecules indicates the
amount of mismatch or base sequence
divergence between the hybridized se-
quences. A /\Tm of 1°C is generally in-
terpreted to indicate about 1% base pair
mismatch (Bonner, et al., 1973).

To measure sequence homology be-
tween single-copy sequences able to form
hybrids under our reaction conditions,
interspecies hybrids were bound to hy-
droxylapatite at 60°C in 0.12 M sodium
phosphate buffer and eluted with 0.075
M sodium phosphate buffer at increasing
temperatures. The elution buffer molarity
was lowered to 0.075 M to ensure that all
DNA eluted from the hydroxylapatite
column with increasing temperature was
the result of thermal denaturation rather
than of a decrease in affinity of the
hydroxylapatite for the double-stranded
DNA (Martinson and Wagenaar, 1977).

Thermal denaturation curves were con-
structed for hybrid molecules formed at
two Co^ values, equivalent Co^ 884 and
equivalent Cot 28,275. A 1000-fold to
2500-fold excess of total A. serenana
DNA at 5 mg/ml was mixed with the
heterologous ^"^H-tracer and homologous
^^C-tracer species, each at 2-5 /xg/ml.
The mixtures were heat-denatured and
reassociated at 66 °C (approximately Tm
— 25°C) in 0.6 M sodium phosphate
buffer. Under these conditions the tracers
are too dilute to interact significantly
with themselves and thus any labeled
hybrids formed are the result of reaction
with the driver species. Most of the
single-copy fractions isolated at equiva-
lent Cot 350 contain some contaminating
repetitive sequences (see Fig. 31, the
homologous reaction component curves).
Precautions were therefore taken to mini-
mize the contribution of contaminating
repetitive sequences to the melting pro-
files since we wished to measure homol-
ogy of single-copy sequences. All repeti-

tive-sequence contaminants should be
reacted at the early Cot point, and any
duplex formation occurring between the
early and late Cot points should there-
fore involve only single-copy sequences.
Thus we have subtracted the hybrid
melting curves of the rapidly reacting
duplexes from those of duplexes formed
at the high Cot value to produce corrected
curves for single-copy duplexes. These
corrected curves were used to determine
values for the ATm between homologous
and heterologous hybrid duplexes.

This correction has a very small effect
on the ATm values observed in Table 4,
even in cases where significant repetitive
sequence contamination is present. In
preliminary experiments wdth total DNA
we observed much lower interspecific
ATm values for repetitive sequences than
for single-copy DNA; similar observa-
tions have been made for Xenopus (Galau
et al, Year Book IJ^^ pp. 711-723), pri-
mates (Deininger and Schmid, 1976) and
Osmundaceous ferns (Stein, unpublished) .

Table 4 summarizes the results of our
experiments. The ATm values range be-
tween 1.5°C and 6.5°C. All but one of
the values are clustered in the 5.5°-6.5°C
range. The exceptional case, with a A^m
of only 1.5°C, involves A. fruticulosa,
which is interfertile with the reference
species A. serenana. The other species,
irrespective of subgenus classification or
photosynthetic mechanism, seem to be
almost equally distinct from .4. serenana.

In our previous experiments using ^^^I
A. hortensis single-copy tracer, we found
a ATm between A. hortensis and .4.
serenana of 8.0°C {Year Book 75, pp.
362-367), compared to the 6.5°C ob-
served with our present techniques. In
the former experiments, the AT„i had to
be estimated by comparing two separate
melting profiles from incubations with
A. hortensis and .4. serenana driver DNA.
We feel that greater accuracy is achieved
with the present procedure, in which two
tracers are reassociated and melted to-
gether with the same driver DNA. ^lore

250

CARNEGIE INSTITUTION

TABLE 4. Summary of Cross Reactivity of Enriched Single-Copy Tracer Sequences with

A. sercnana Total DNA Driver Sequences*

Sinsile-Copv

STm at Equiv

alent Cot

Photosvn

thet

ic

N

ormalized -

Corrected

Tracer

Pathw

■^y

Subgenus

x-Rxn

884

28,275

AT„»

A. fruticulosa

a

Obione

0.85

1.5

2.5

1.5

A. sabulosa

a

Euatriplex

0.65

4.5

4.75

5.0

A. rosea

C4

Euatriplex

5.75

t

t

A. hortensis

a

Euatriplex

0.45

6.5

6.5

5.25

A. truingularis

a

Euatriplex

0.51

5.0

6.5

6.25

A. phi/Uostcgia

c.

Obione

6.5

t

t

* The extent of cross reactivity is normalized with respect to the reactivity observed for
A. screrwJia tracer in each reaction. The ATm is the difference between the Tm values for
homologous and heterologous hybrid molecules in each reaction. The corrected ATm is derived
from the curve of the difference between high and low Cot thermal denaturations.

t Data for the difference denaturation profiles was not reliable because thermal degradation
was observed at high Cot values. However, the homologous tracers in these reactions were
relatively free of repetitive-sequence contaminants. Therefore, low Cot ATm values are as-
sumed to represent an accurate ATm for comparative purposes.

importantly, the use of 0.075 M rather
than 0.12 M sodium phosphate buffer
probably gives a much more realistic
estimate of strand separation (as opposed
to duplex elution) in these experiments
(Martinson and Wagenaar, 1977).

The extent of cross reactivity between
each single-copy tracer and A. serenana
DXA was also measured by comparison
of the heterologous and homologous re-
association kinetics. Mixtures were made
up and incubated as for the thermal
denaturation curves. At increasing Cot
values, aliquots of each mixture were
chilled, diluted into 0.12 M sodium
phosphate, and fractionated on hydrox-
ylapatite at 60°C. Figure 31 shows the
results of the hybridization of A. serenana
total DNA with the single-copy tracers

from A. hortensis and A. fruticulosa.
Theoretical second-order components
have been fitted to the data for the A.
serenana self-reactions by a least-squares
procedure, using a computer program
supplied by Dr. Eric Davidson and
adapted to our computer by Mr. Glenn
Ford.

Mismatched single-copy heteroduplexes
form at reduced rates, relative to the
rate of A. serenana self-reassociation.
Heterologous reactions may therefore be
incomplete at the highest attainable Cot
values, simply because no unreassociated
A. serenana DNA remains to drive fur-
ther reaction of the tracer. To estimate
potential cross reactivity in such cases
we have fitted the heterologous reaction
data using the expression

G/Go = J + F, n + KmCot)-^

+ F.,

(l+Ko2Cot) -'■'''' exp

/ 0.25T\:i-{l+Kr>2Cot)^^^] \
\ 056 /

G^Gf, is the fraction of the tracer re- rate constants of Koi and Ko2j respec-

maining single-stranded at a given Cot, tively. T is the retardation coefficient

J is the fraction incapable of reaction, (i^2. tracer/^/) 2) estimated from the ex-

F, and Fo are the fractions of the tracer perimentally determined AT^ of the

reacting with driver components having heterologous duplexes using the data of

DEPARTMENT OF PLANT BIOLOGY 251

Bonner et al. (1973). Repetitive se- feel that the lar^e differences in DNA

quences contaminating the tracer prepa- homology previously observed between,

rations are assumed to react at the same for instanc(;, A. hortensis and either A.

rate (Kdi) as the most slowly reassoci- triangularis or A. sabulosa, make this

ating driver repeats. The expression for hy[)othesis much more complicated and

the retarded reassociation kinetics of the less tenable.

single-copy component was derived by Support for the assumption that ge-
Galau et al. (Year Book 74, pp. 711- nome sizes do differ significantly among
723).^ It permits us to extrapolate from different Atriplex species can be derived
the observed data to estimate the theo- from a comparison of the rate of single-
retical potential cross reactivity for copy sequence reassociation observed
various tracers. Examples of such extra- previously for A. hortensis with that ob-
polated fits are shown in Fig. 31. Table 4 served here for A. serenana. The A.
gives the fraction of single-copy tracer serenana rate is significantly faster than
sequences capable of cross reacting with that for A. hortensis, even after correct-
single-copy sequences of A. serenana for ing for the effects of the different buffer
the four cases in which suitable kinetic systems used in the two sets of experi-
data are presently available. ments, which suggests that the A. serenana

Cross reactivity was highest with A. genome is substantially smaller than that

jruticulosa, as was expected from the of A. hortensis. Further experiments are

hybrid thermal stability and genetic planned to extend this analysis to other

hybridization results mentioned above, species.

Significantly lower values were obtained The new data presented here for both

in reactions with the other tracers; 35- cross reactivity and hybrid thermal sta-

55% of the single-copy sequences in these bility support our previous conclusion

three Euatriplex species were apparently that the main species groups in Atriplex

unable to react with sequences in the are quite distinct from one another and

A. serenana genome under our incuba- that lines leading to the various groups

tion conditions. In the case of A. /lor^ensis, probably originated from an ancestral

the 45% cross reaction observed here stock at about the same time. Our data

agrees well with the value of 48% we are clearly inconsistent with any scheme

obtained previously (Year Book 75, pp. such as that of Hall and Clements (1923)

362-367). The iodinated A. hortensis in which separation of the Euatriplex and

tracer used in the earlier experiments Obione subgenera is assumed to have

contained about 5% repeated sequences, occurred much earlier than divergence of

which probably contributed to the total the lines leading to the main species

cross reaction. groups within each subgenus. We there-

We are assuming that the DNA which fore suggest that conclusions based on
fails to cross-react with A. serenana DNA such schemes should be carefully re-
is composed of sequences which were evaluated. Although we cannot exclude
eliminated in the evolution of present- the possibility that C^ photosynthesis
day A. serenana away from the main evolved independently in two or more
generic branch. Addition of new DNA lines, our data suggest that any such
sequences to the other species groups lineages must have had their origin at
(e.g., by hybridization with a species in nearly the same time. For the present, it
another genus) is also possible, but we appears simpler to assume that several

lines leading to the contemporary C4
species groups separated from a single C4

iT^v.^ „ +-^ • + J • +1 • 4.U stock early in the historv of the genus, at

I he equation was printed incorrectly in the '^ . -^ . ^

original publication. Dr. Roy Britten has con- about the tmie the other major evolu-

firmed that the one given here is correct. tionary lines originated.

252

CARNEGIE INSTITUTION

References Deininger, P. L., and C. W. Schmid,

Bonner. T. L. D. J. Brenner, B. R. Xeu- Scitmce 194, 346-348, 1976.

ieklaudR. J. Britten, J. Mol. Biol 81, Hall, H. M., and F. E. Clements, Car-

123-135, 1973. negie Inst. Wash. Publ. 826, 1923.

Britten, R. J.. D. E. Graham, and B. R. Martinson, G. G., and E. B. Wagenaar,

Xeufeld. Methods Enzi/moi :39, 363- Biochim. Biophys. Acta 474, 445-455,

418, 1974. ^ 1977.

INTERSPECIFIC HYBRIDIZATION OF FERN DNA

Diana B. Steiji^ and William F. Thompson

Osmunda regalis, 0. claytoniana, and
0. cinnamomea are three ferns whose
taxonomic relationship has been a source
of controversy for almost 70 years. We
have been studying relationships among
these species by means of DNA sequence
comparisons. Analysis of interspecific
duplexes formed with labeled 0. clay-
toniana DNA originally suggested that
this species was more closely related to
0. cinnamornea than to 0. regalis (Stein
and Thompson, 1975; Year Book 74, pp.
786-789 1 . This conclusion was based on
several measurements of the number and
thermal stability of duplexes formed
under a wide variety of reassociation
conditions. However, all these experi-
ments used DNA prepared from a single
collection of fronds of each species, and
subsequent experiments with DNA from
different collections showed less homol-
og>' between 0. claytoniana and 0. cin-
namomea than had been observed ini-
tially. We have attempted to identify the
source of this discrepancy and to deter-
mine the range of variation in the results
of inter.'^pecific hybridizations between
different DNA preparations or between
tissue samples; in our experiments DNA
samples obtained from separate fern

^ Pro5!ont addrf'.s.s: Dfpartmont of Zoology,
University of Ma.ssarhusot.ts, Amhfr.st, Massa-
chu.setts 01003.

populations at different seasons in four
different years were used.

DNA was isolated, sheared, and iodi-
nated as previously described (Stein and
Thompson, 1975; Year Book 74, pp. 786-
789). 0. claytoniana DNA was used as
the tracer DNA in these experiments,
since its position relative to the other
two species was in question. Aliquots of
labeled DNA were mixed with a 1,000-
fold to 5,000-fold excess of DNA from
0. cinnamomea, 0. regalis, or 0. clay-
toniana; the mixtures were denatured
and incubated at 42° in 50% formamide
containing 1.0 M NaCl and 10 mm Pipes
buffer, pR 6.7-6.9 to a Cot of 2000. At
this Cot virtually all repeated sequences
(but few single-copy sequences) have
become reassociated. Duplexes were then
adsorbed to hydroxy lapatite in 0.12 M
sodium phosphate buffer and subjected to
thermal elution.

A typical thermal elution profile ob-
tained with the 1975 DNA's is shown in
Fig. 32. The T^ (temperature at which
50% of the adsorbed DNA has been
eluted) reflects the average precision of
base pairing in a given population of
duplexes, and the AT^, or difference in
7\n between inter- and intra-specific
duf)lcxes, provides an indication of the
flegree of sequence divergence for a given
[)air of species. The results of eight such

DEPARTMENT OF PLANT BIOLOGY

253

r — I 1 I 1

\

—^f—

r

—

100

-o .A^

■l: " "

■^

T3

-• ' ' ■•^^'^

<^

'^'•'' X

D

80

/y^ /

-

Q)

/. ' ' /

.^

*•' /

C

f / /

o

l_

60

-

0)

/." / /

a

/'■ / /

(^

/• / /

>

40

.'■ rl J

-

-^ ■

' y^ 9

/■ / r

3

E

jf / /

3

20

>f^^^^^ 1^ 1 1 1

1

—'—ii—

1

65 70 75 80 85 90 95 0.4MNaPB

Temperature, C

Fig. 32. Thermal elution profiles for sheared native 0. claytoniaua DNA (•) and duplexes
formed between 0. claytoniana ^^T tracer and unlabeled DNA from 0. claytoniana (O), 0.
cinnamomea (A) or 0. regalis (□) (1974 preparations of DNA). Except in the case of native
DNA, appropriate samples were heat-denatured and incubated as described in the text. Dupli-
cate samples for each species were then diluted 200-fold with 0.12 M sodium phosphate buffer
(pH 6.8), sonicated briefly, and applied to hydroxylapatite at 60°C. After the columns were
washed to remove single-stranded DNA, thermal elution was performed by raising the tem-
perature of the column in increments of 5°C and eluting with 0.12 M sodium phosphate buffer
at each temperature. Results are presented as the cumulative percentages of initially bound
DNA eluted at each temperature.

experiments are summarized in Table 5
in terms of ATm values. The data show
that, with the exception of the samples
from the 1973 collections, DNA's from
0. regalis and 0. cinnamomea appear to
be about equally diverged from that of
0. claytoniana. Variation between experi-
ments utilizing DNA's from one extrac-
tion is small, as is the variation observed
from year to year — with the exception
of 1973.

Several variables in experimental pro-
cedure were examined in an attempt to
explain the apparent close homology be-
tween 0. cinnamomea and 0. claytoniana
observed in the 1973 collections. These
variables included sonication of the
duplexes to disrupt any aggregates before
application to hydroxylapatite, iodina-
tion of DNA fragments in double-
stranded versus single-stranded form,
different ratios of tracer to driver DNA,

and possible fragment length effects.
None of these factors produced signifi-
cant changes in AT^ values.

To ascertain whether it was 0. clay-
toniana or 0. cinnamomea DNA that
behaved differently in the 1973 prepara-
tions, 1975 0. claytoniana tracer was
reassociated separately with 1973 and
1975 0. cinnamomea DNA. Nearly iden-
tical thermal elution profiles were ob-
tained for both hybrid populations (Fig.
33). The AT„, was about 4.2° in both
cases, which is within the range of values
obtained for all other experiments with
post- 1973 preparations. When 1973 0.
claytoniana DNA was used to prepare
the tracer for a similar experiment with
the same two preparations of 0. cinna-
momea driver, it was again observed
that the hybrids formed with the two
driver DNA's had nearly identical ther-
mal elution profiles. However, in this

254

CARNEGIE INSTITUTION

TABLE 5. Comparison of Data from Experiments with ^"''I-labeled 0. claytoniana Tracer

under Standard Conditions*

Collection Date

lodinationj

ATm for Hybrids with

0. cinna}7Wt)ica

0. rcgalis

Sept. 1973t
August 1974

>ept. 1975

June 1976

Mrnn ± -tandard deviation (1974-1976)

DS§

ss§
ss
ss
ss

DS

SS

ss
ss

1.5X

3.6

4.5

3.8

3.8

3.7

4.0

4.5

2.8°C

3.8

4.0

3.8

3.5

3.2

4.0

3.5

4.01 ±0.35

3.69 ± 0.30

* Incubation in 507c formamide, 1 M NaCl, 0.01 M Pipes (pH 6.7-6.9) at 42° to Cot =
2.000.

t A large number of experiments involving various reassociation conditions were carried out
with the. 1973 DNA; although absolute ATm values varied with the different conditions, results
were always qualitatively the same.

t lodination conducted using double-stranded (DS) or single-stranded (SS) DNA.

§ In these experiments, reassociated duplexes were sonicated briefly to disrupt large hyper-
polymers before being apphed to hydroxylapatite.

il 1973 0. cinnamomea driver was used in this experiment.

100 -

no

•*-

Q)

C
O

u

«»

a

>

E

65 70 75 80 85 90 95 0.4MNaPB

Temperature, °C

Fig. .33. Thermal elution profiles for sheared native 0. claytoniana DNA (•) and duplexes
formed between 1975 0. cl/jytoni/ma ^^T tracer and unlabeled DNA from 1975 preparations of
O. cUiytoni/ina (O), 0. cinnamomea (A), 0. regalis (\J), and the 1973 preparation of 0.
cinnamomea (V). Experimental method5 were as described for Fig. 32 except that reassociated
DNA'.s were not .sonicated before application to hydroxylapatite. Triplicate samples represented
by a single point when values were the same. <x indicates coincidence of data points for the
two preparations from different years.

DEPARTMENT OF PLANT BIOLOGY

255

case the l\Tm was only about 2.2°,
which is close to the value ol)served in
the 1973 experiments. The /\Tm meas-
ured with 1975 0. regalis driver was
nearly the same (3.5 vs 4.0°) whether
1973 or 1975 0. claytoniana tracer was
used. Thus it was possible to reproduce
the 1973 results merely by substituting
1973 0. claytoniana tracer in experiments
with the later preparations, and we con-
clude that the 0. claytoniana prepara-
tion was primarily responsible for the
difference between our results with 1973
and results with later preparations.

This conclusion was confirmed by an
experiment designed to compare the 1973
and 1975 0. claytoniana preparations
directly. DNA from each of the two
preparations was iodinated in parallel
reactions and the resulting tracers tested
for their ability to reassociate with un-
labeled 0. claytoniana DNA from 1975.
After incubation to Cot 2,000, 76% of
the 1973 tracer and 86% of the 1975
tracer had reacted. Too little of the 1973
tracer was left for its reactivity with the
1973 driver to be measured, and thus we
do not have a rigorous control for this
experiment. However, we have observed
between 85% and 89% self-reassociation
in all previous experiments. We therefore
believe that the lower cross reactivity
observed with the 1973 tracer indicates
that the 1973 0. claytoniana preparation
contained sequences which were not
present in later preparations from this
species.

Since the 1973 0. claytoniana DNA
also showed more homology with 0. cin-
namomea than did later preparations, it
is reasonable to assume that some se-
quences from the 0. cinnamomea genome

were present in this DNA. This circum-
stance could be explained in either of
two ways. Some 0. cinnamomea fronds
might have been inadvertently mixed in
with the 0. claytoniana collection that
year. (These two ferns are so similar in
appearance that fertile fronds must be
present for positive identification.) Al-
ternatively, it is possible that the sup-
posed 0. claytoniana material collected
in 1973 came from a population of 0.
claytoniana X 0. cinnamomea hybrids.
Such putative hybrid populations are
currently being investigated by H. Ahles
(personal communication). There is evi-
dence that hybridization does occur
within this group, in the reported occur-
rence of a hybrid between 0. regalis and
0. claytoniana var. Rugii (Try on,* 1940).
It is noteworthy that, with the single
exception of the 1973 0. claytoniana
preparation, we have seen very little
variation in results from different ex-
periments, including ones using different
DNA preparations and tissue collections.
The sensitivity of these comparisons thus
appears to be quite high. Since the 1973
0. claytoniana DNA probably originated
either from a natural hybrid population
or from an inadvertently constructed
"hybrid" mixture of tissue, the repro-
ducible differences between this and sub-
sequent preparations suggest that molec-
ular hybridization techniques may be
sensitive enough to permit future studies
of natural hybrids, and perhaps even
races or sibling species.

References

Stein, D. B., and W. F. Thompson. Sci-
ence 189, 888-890, 1975.
Tryon, R., Amer. Fern J. 30, 65-66, 1940.

CONTAMINANTS AFFECTING PLANT DNA REASSOCIATION

Michael G. Murray and William F. Thompson

Studies of DNA reassociation, sequence partly because it has been harder to ob-
organization, and evolution have been tain sufficiently pure plant DNA in the
more difficult in plants than in animals, required quantities. Most plant tissues

256

CARNEGIE INSTITUTION

contain low concentrations of DNA and
high concentrations of materials such as
tannins and soluble polysaccharides
which are difficult to separate from DNA
by conventional techniques. Recent prog-
ress in plant genome organization has
been greatly facilitated by development
of new isolation procedures. An essential
step in most such procedures involves the
selective binding of DNA to hydrox-
ylajxitite in the presence of 83/ urea and
0.24 -)/ sodium phosphate buffer, with or
without other additives such as NaC104,
SDS, EDTA. or 2-mercaptoethanol
(Britten et a/.. 1974). RNA, protein,
tannins, and the majority of cellular
polysaccharides can usually be removed
by washing the hydroxylapatite with
urea-phosphate buffer prior to eluting the
DNA with 0.4 or 0.5 M sodium phosphate.
Hydroxylaj^atite chromatography alone
can yield DNA that appears to be pure
by the standard criteria such as ultra-
violet absorption spectra, thermal dena-
turation profiles in standard buffer sys-
tems, and analytical CsCl gradients.
However, as normally applied all these

Fig. 34. Schlieren photograph of an over-
loaded analytical CsCl gradient containing pea
DNA purified by hydroxylapatite chromatog-
raphy, RXa.se and prona.se digest ion.s, followed
by chloroform/octanol extraction and ethanol
precipitation. A .solution containing 1.89 mg/ml
DX.\ in 10 mM .sodium acetate, pH 7.0 wa.s
adjusted to an initial den.sity of 1.696 g/cc with
.solid CsCl and centrifuged at 44,700 rpm for
44 hr at 21 "C in an AN-D rotor using a 12-mm
single sector cell. Precipitated contaminants
formed an opaque band above the DNA caus-
ing the dark vertical line in the photograph.

criteria are sensitive only to contaminants
that have significant absorption in the
tiltraviolet. We have recently observed
that plant DNA prepared by the hy-
droxylapatite procedure can be signifi-
cantly contamined by polysaccharide-
like impurities even though they appear
pure by these standard criteria. Figure
34 shows the schlieren profile obtained
from an overloaded analytical CsCl
gradient containing pea DNA purified
by hydroxylapatite chromatography. In
addition to the typical schlieren pattern
of the DNA band, we observe a vertical
line in the photograph corresponding to
precipitated material of a slightly lower
density than the DNA. This precipitate
band could be observed visually in the
cell. Its density suggests that it is com-
posed principally of polysaccharide (s),
although as yet we h